Home > Learn > Digital Post-Processing > Fine Art Photography Workflow
Digital photography requires a solid workflow, allowing for professional preparing of digital photo files for the web and print. For the Digital Photography Workflow series, we consulted with a number of experienced professional photographers who are also stellar photo.net members and frequent contributors to the Photo.net Digital Darkroom forum, to walk us through their specific digital photography workflow.
In this article, Michael Ezra discusses his unique digital workflow process tailored to meet the needs of his professional fine art photography career, the set of software and tools he prefers to use, and goals he accomplishes with his digital workflow. The article is enhanced with illustrative figures and screen shots, and includes examples of fine art photography from Michael Ezra's portfolio. Whether you are just entering the world of digital photography and needs some tips and advice on how best to post-process your images, or are a seasoned pro, the insights shared here should be helpful with your own digital photography workflow and fine art photography post-processing.
Effectiveness and efficiency are often hailed as the defining corner stones and driving forces behind the progress of the modern age. This holds especially true for the world of photography, particularly for its execution and processing aspects—an invisible mechanism behind the final works of art. Introduction of digital technology into photography had a double-faced effect on the field. On one hand, it undoubtedly brought some groundbreaking changes to the effectiveness of the photography as a complete art medium. On the other hand, its contribution to efficiency may be considered less certain given that by introducing better tools digital technology unlocked a flood of new possibilities—a flood that often threatens to overwhelm even a more experienced photographer.
This article comes as a result of years invested in researching and analyzing the ways to do photography better; not the artistic fine-tuning, but the actual tools of the trade -technologies underlying digital cameras and backs, powers of RAW processing, pitfalls of post-processing, preparation of output for printing, etc. My motto is simple: efficient process allows concentrating on art while proper processing allows bringing that art to life; and I offer this article to those readers who are looking to explore the possibilities offered by digital technology in the studio workflow and to master the aspects important for achieving a greater efficiency in this field to propel their progress.
Solving the puzzle of establishing an efficient photographic studio workflow is similar to solving any other problem for process flow: it is possible to arrive to a solution efficiently only when requirements for the end result are well defined. Envisioning and understanding the end results—in our case, it can be fine art prints in a particular size, images for the web galleries, etc.—drives the selection of proper tools while the sequence and efficiency of their usage comes with experience. Bottom line: a good workflow should allow achieving desired results repeatedly, reliably, and with consistent quality.
In this article I would like to overview and share my studio workflow specific to producing fine art photographic prints of studio nudes. From the technical point of view, the utmost important aspect of the end result in production of fine art prints is their quality. Fine art prints should truly be very fine; therefore attention to detail in every step of the process is critical. My workflow is tuned specifically for this aspect as the highest priority.
From A to Z, the workflow is defined by the following key elements:
As we go down the list, I will describe each key element individually and also in relationship with the other elements. But before we begin, here is a very concise look at the entire workflow (Figure 1: Photography Workflow). This diagram lists all steps highlighting when and what kind of results are produced, along with recommendations for storage and back-up, and it can be used as a quick reference as you move through the article.
The preparatory steps described in this section do not necessarily take place during every single photoshoot. As a matter of fact, a lot of the decisions here are made once, and then they become just another integral part of your routine.
Studio space is customized based on individual project requirements. This includes selection of specific backgrounds, lighting support, props, as well as setting of a comfortable room temperature. Remove all unnecessary objects in order to free the working space and to be able to move unhindered, especially in the capturing area.
The primary lighting setup in my studio consists of a set of strobe lights (constant lighting would not work because models are not perfectly still subjects, and the smallest shake or movement will affect the sharpness of the final image). Personally, I prefer using monolight strobe units for the following reasons: they are independent light sources and accidental breaking of any one of them would not influence the others; each monolight can provide a substantial light output; they require less cabling. Depending on the image in mind, each of the light sources can be individually set to specific power levels and can be outfitted with light modifiers, such as softboxes, beauty dishes, grids, color filters, etc. Reflectors can also be used to assist in proper distribution of light in the studio and on the subject and to achieve the desired light painting effect. I found proportional modeling lights very helpful in previewing the light pattern, which will be produced with flash lighting. A traditional light meter can be used to measure the light distribution in the setup but nowadays digital cameras provide an easier way to preview the results of the lighting setup, substantially simplifying the task of achieving desired lighting.
Monolights are usually equipped with photo slaves. Therefore, if at least one flash fires others will immediately respond in unison. If simultaneous firing of all flashes works for your projects - like it does in my case - then the setup is very straight-forward. You can trigger the "leading" flash via either a wireless trigger or a cord. If using a cord, just connect it to the flash unit closest to you to minimize tangling and to keep it out of the camera view.
For the purposes of this article, we will focus on digital technology. Any professional-level digital photo studio should have in its arsenal two or more cameras. Technical difficulties do happen with even the best-in-breed cameras, and they should not be show stoppers. My own experience taught me that having a back-up camera is not just beneficial but can be a lifesaver.
If your priority is the ultimate image quality and not necessarily the speed of capturing, a medium format digital camera or back would typically be a better choice for your studio. They are slower but produce images of superb quality. In my studio, I use a Mamiya ZD medium format digital camera (Figure 2: Photography Equipment).
645 System
35mm System
6x7 System
Lighting Equipment
Printing System
Computer System Tools and Utilities
Regardless of the format of the camera you choose to use, prior to the photoshoot you should inspect its image sensor to determine if cleaning is required. To do so, set a lens aperture to the highest setting (typically 22), point the camera to a white image displayed in full screen on a clean computer monitor, defocus the lens, and take a picture. Review the picture on the computer monitor at 100% zoom. What you see is what you will get in every image captured (dust specs will appear softer with wider aperture settings). When deciding to clean or not to clean, be aware that opening the camera body always carries a risk of introducing additional dust specs. If you decide that cleaning is not necessary even though some dust specs are present, still save the dust reference image as it can be used either for faster manual locating of dust spots in the subsequent images or for feeding into dust removal software (some cameras have this feature built-in). If cleaning is required, follow the manufacturer's guidelines.
The choice of a lens is determined by the desired field of view, the available studio space, and the nature of your project. Focal length of the lens affects depth of field, focusing distance, and visible proportions of the objects at various distances from the lens. Fixed focal length lenses are generally the sharpest, while quality zooms offer quick versatility. Speaking about quality, it is important to realize that all lens manufacturers have their great, good, and not-so-good lenses. Hence, it is important to know specific performance characteristics (and weaknesses) of the lens when deciding on using it (or even better, purchasing!). There are several great online resources that can help you in this regard.
I recommend carrying out studio capturing with a camera set to either manual mode or, if present, a special "X" mode, designated for photographing with strobe lights. In either case, the shutter speed should be set to the highest value of the sync speed supported by both the camera and light units. Take a test image: if any part of the frame looks darkened, it may indicate an incorrect setting of the shutter speed as shutter itself is being captured in the image. If you are using a wireless trigger, the darkening can also be caused by the latency in the triggering mechanism. To troubleshoot this, trigger with a cord.
If your project requires freezing a motion (e.g. jumps), then you should go for strobe lights with a shorter flash duration and a camera system with a higher sync speed. In all other cases, a more or less standard sync speed of 1/125 of a second is sufficient. Shorter flash duration will always lead to sharper images. But keep in mind that on some light units shorter flash durations are achieved by reducing their power output (less light).
After you lock the shutter speed, you can control the exposure setting of the camera with the aperture dial. Considering that aperture influences the depth of field, I first decide on the aperture based on the look of the image I intend to capture and then adjust the power setting of the strobe lights to provide a correct exposure with the selected aperture.
A camera's LCD display can provide a quick and informative feedback on the capturing conditions. I find the histogram to be the most important feedback of the digital camera as it allows to analyze exposure conditions in detail. While some cameras provide individual histograms for R, G and B channels, the majority display the luminance histogram, which is derived from the relationship of captured R, G and B channels—a white balance. Thus, a correct setting of the camera's white balance prior to capturing makes the displayed histogram meaningful and, thus, has the direct impact on the accuracy of photographer's judgment on exposure. In the studio workflow, I recommend to measure and set the custom while balance for each photo session using a gray card (follow camera-specific manufacturer's guidelines).
In addition to histogram display, some cameras are equipped with a feature of blinking those areas of the preview in the captured image that exceed threshold levels—those which are over- and under-exposed.
The exposure settings during capture have a direct impact on an unavoidable and thus important factor—image noise. Noise inherent to a digital image is introduced in two stages: during the image capture process and during image recording process.
Firstly, you can minimize the capture noise by increasing the amount of light being captured. This can be achieved by a) increasing the light output of the strobe lights; b) opening the aperture of the lens to allow more light in; c) increasing duration of capture (not really applicable when strobe lights are used). Considering that studio environment provides full control fover the light used for image capture this part is usually not an issue.
Secondly, you need to ensure that your camera is tuned to measure/record the maximum information. Keeping the camera at its lowest ISO setting will enable it to operate at the maximum of its capability. This setting will permit the camera to capture the widest possible range of brightness in the scene and, subsequently, record it with the least amount of noise.
Due to specific design of the image sensors, the medium format digital cameras are particularly advanced in capturing wide dynamic range. The only lower-end professional DSLRs that can currently compete with them in this territory are Fuji S3 and Fuji S5 models. This excellent wide dynamic range capability of medium format digital cameras in the studio environment can be put to a good use through purposeful careful overexposure of the captured images by about 1-1.3 f-stops in order to ensure the capture of the cleanest files possible, with noise-free shadows. Subsequently, during development of raw images you can apply a compensating negative exposure correction to straighten the brightness of the image files.
Photography, even digital, is all about light—capturing it and then recording it. While previous section discussed how to optimize the capture aspects, here I would like to say a few words about successful image recording.
The accuracy with which captured light is recorded into a digital file is characterized by a) the spatial resolution of the file (measured in megapixels) and b) the precision of digital encoding of the color and brightness of the captured signal. Loss of precision during the recoding process leads to loss of the image data. Different cameras offer different flexibility when it comes to image recording parameters. My simple recommendation is to always use the full resolution of the camera's sensor and the highest available settings for recording color and brightness nuances. In other words, capture in RAW format exclusively! (see Table 1)
| Setting | Recommended Value |
| Image format | RAW (uncompressed or with lossless compression); plus optionally small JPG preview in high compression, unless camera is slowed significantly |
| Resolution | Maximum |
| Color space | Widest available (irrelevant for RAW) |
| Bit depth | Maximum, if selectable |
| ISO | Native ISO, usually minimum value |
| Shooting mode | Manual or "X" |
| Shutter speed | Fixed at camera's sync speed |
| Aperture | Set based on desired depth of field; controls exposure |
| Exposure compensation | Off |
| Image rotation | On |
| Sharpening | Standard (irrelevant for RAW) |
| Image tone | Standard (irrelevant for RAW) |
| Noise reduction | Off |
| Allow capture without a memory card | No |
| Image capture method | Tethered or to memory card(s) |
One of the important conveniences of the studio setting is that recording of captured images can be extended to their essential destination—a computer. Instead of recording images on memory cards, they can be transferred directly to computer storage as they are being captured, either wirelessly (for some camera models) or using Firewire or USB cable connections. Some cameras, Mamiya ZD for example, have a great feature: if writing of image data to a computer fails, they can automatically switch to writing to a memory card instead so images would not get lost. This is a valuable feature as cable connections are not 100% reliable.
Capturing tethered to a computer has a number of advantages, such as a large size accurate image preview on a calibrated computer monitor (which can also turn into a distraction), a more detailed image histogram, a more secure storage media, automatic naming and proper placement of the image files, and time saving on a subsequent bulk data transfer from the memory card to the computer storage. Capturing to a memory card, however, has an advantage of a completely wireless camera system (when wireless sync with strobes is used). During my photo sessions I often practice both: I usually start with tethered capturing and later on, when lighting is fully confirmed and if greater mobility is required, I may switch to memory cards. (Figure 4: Mamiya Digital Photo Studio, tethered capturing)
Faster and reliable memory cards are certainly preferred; their important characteristic for studio use is the ability to withstand accidental falling on a floor, as shock can lead to data corruption. So far I have had no problems with SanDisk Extreme III and IV series 4 GB and 8 GB cards. Luckily the drop test has not yet been conducted in my studio. Extreme series cards are usually packaged with file recovery software that can be very handy in other more creative accidents.
In order to assure that the colors and brightness displayed by the computer monitor are accurate, the monitor must be properly calibrated. Use of the calibrated and profiled monitor is absolutely critical at every step of the photographic workflow, starting from the tethered image recording when images are displayed on the screen to provide visual feedback, certainly during subsequent editing of the images and at a later stage, while preparing them for printing. An uncalibrated monitor will introduce major discrepancies in image data display and will be a killer for digital workflow.
It is important to understand the difference between calibration and profiling. Monitor calibration is a process of adjusting internal characteristics of the monitor itself. Monitor profiling is essentially calibration of the video card output, i.e. instructing the card to send proper signals to a calibrated monitor. Only some monitors on the market can be calibrated by a user. Most of the monitors are factory calibrated, but allow adjustments of brightness, contrast and color enabling a user to fine-tune the monitor calibration during the profiling procedure. Most software packages handle calibration of the monitor only within these limits, as they are mostly designed for video card profiling.
Until a very recent upgrade to an NEC 2690WUXi LCD, I was using a Mitsubishi 2070SB CRT analog monitor along with Gretag Macbeth Eye One Display calibration unit. My first generation Eye One Display unit simply stopped working in about a year. The second generation Eye One Display 2 seems more reliable—it is still operational and provides good calibration and profiling results. Eye One Display 2 is supported by various monitor calibration and profiling software packages, but I found the software that it comes with—Eye One Match—to be the most optimal solution. Eye One Match offers a very limited support for Display Data Chanel (DDC), which enables a fully automated software control over the monitor settings, and, hence, all monitor adjustments have to be made manually. Though manual process is tedious, calibration results are quite good. One of the ways to measure the quality of the achieved monitor calibration and profile is to evaluate the grayscale gradient in a color-managed application (such as Photoshop) for smoothness, neutrality, and clipping. Other third-party calibration software packages can additionally provide a quantitative measure of the quality of the achieved calibration result. Yet, while trying them out I ran into compatibility issues and stayed with Eye One Match solution.
As for a recent upgrade, I am still evaluating the SpectraView software from NEC, which allows for fully automated monitor calibration and profiling using DDC.
Considering that the main storage of your computer system is responsible for storing all image assets of your photo studio, it is crucial that it should be able to withstand failures of individual hard drives. If hard disks never failed on you yet, it is pure luck—and it will eventually run out.
There is a variety of storage redundancy solutions available on both a consumer and professional level. One of the simplest redundant storage setups is mirroring of a drive, which is RAID 1. In such a setup each drive has an identical copy created in real-time. The next level with increased storage performance characteristics is RAID 5 configuration, requiring at least 3 disks. This setup can also survive failure of any single disk, but it provides a better read performance because stored data is stripped through multiple disks and can be read in parallel. Initially I chose RAID 5 for organization of the main storage in my studio, until I had a simultaneous failure of 2 disks. Luckily, I was able to recover one of them on another machine, using specialized software, which allowed me to bring the entire RAID 5 online and create a full latest backup.
This incident may seem like a very unlikely event, until one considers the fact that all disks in this RAID 5 array were acquired in a single purchase from a single vendor. As a result, they all were from the same batch and had the same manufacture date. Since they were under basically identical use in the same RAID array, they failed on the same date as well! Thus, if you are serious about building a reliable storage, I highly recommend purchasing disks individually from different vendors. This will also ensure that any possible problems caused by shipping would not influence all disks in the same way.
Mindful of the time wasted for those recovery efforts, I chose to store data at an even higher level of protection, which is RAID 10. In this configuration when 4 disks are used, failure of any 2 disks (unless they are paired in internal RAID 0 stripe) will still not fail the entire storage array. The very high availability of this setup comes with a cost: it is not efficient in utilizing storage space of the disks, with only 50% of the total space of drives being available for usage. RAID 10 provides excellent redundancy and high performance on both read and write operations.
As a final note on this topic, keep in mind that the most common cause of hard disk failure is overheating. Setups with many hard disks require reliable cooling.
The accumulation of image data in digital workflow can be overwhelming. Staying organized is a time saver and the only way to achieve it is to never let the chaos get in! Data should be organized as soon as it is being created.
There is a multitude of software packages available that provide a "user-friendly" organization of digital content. Such solutions are typically based on introduction of an additional layer, such as a custom database, or a catalog, specific to a particular software package. Switching from one such solution to another would involve the painful process of migrating this organizational data from one proprietary format to another, which usually is not a viable task. This imposes a forced loyalty on a photographer.
To avoid this lockup I chose to utilize the file system structure itself as the primary means of organization of all digital content produced in my studio. After all, file system is inherent to storing files and what can be better than having it organized in the first place!
The systematization and organization of photographic data becomes easy once you identify parameters that describe the content and nature of the image files and their collections in a meaningful way. Pertinent to my work with studio nudes, I employ the following hierarchy of parameters: the type of imaging (Imaging_Nudes_Studio), subject (a particular model or a project name), date of capture, etc. (See Figure 5 and Table 2). It is a very simple but descriptive system, which allows for great flexibility.
| Folder Name | Notes |
| Imaging_Nudes_Studio_NU | Contains all files specific to nudes shot in the studio |
| Models FullName_UniqueAbbreviation | Contains all Studio Nudes files for a particular
Model Unique abbreviation constructed from model's name will be used in naming all files created with this model. When multiple models are shot together, the unique abbreviation is extended. Abbreviation should be at least 3 characters long. Example: Jane_Doe_JDE |
| PhotoshootDate_AlphaIndex | Contains all Studio Nudes files for a particular
Model for a particular date of the photoshoot Format is YYYY-MM-DD_A Alpha index is the sequence index of the photoshoot; it will be carried over to the filenames in this folder. Example: 2006-01-31_A 2007-02-15_B 2008-03-21_C |
| 1 - Raw | Contains Raw image files |
| 1.1 - InProcess | Contains files that are being edited |
| 2 - Previews | Contains low resolution JPG previews |
| 3 - Contact sheets | Contains "contact sheets" generated from preview JPG's |
| 4 - Finished Works | Contains Master files and derived works (prints, web, etc.) |
| 5 - Prints for Model | Contains print images made specifically for a model (if any) |
| 6 - Documents_Notes | Contains various documents related to a particular photoshoot, such as scanned copy of a signed model release, copy of the model's ID card proving age, scanned receipt of payment, text documents with any special notes. Most of these documents can be consolidated into a single PDF file. |
The hierarchy of data organization travels into my file naming convention, so that the name of each individual file is comprised of elements representing each level of its parent directory structure. In order to keep the filenames short, only the unique abbreviations and indexes from the directory names are used when filenames are constructed. In the example below, a sample filename for the RAW file is constructed from the bolded parts only:
Imaging_Nudes_Studio_NU_John_Doe_JDE_2006-01-31_A_SequenceNumber.FileExtention
The result will be the raw files named as follows:
NU_JDE_A_001.raw,
NU_JDE_A_002.raw,
etc.
Development of the RAW file into a TIFF or a PSD file results in the creation of the Master file for that particular raw file. During photographic workflow, Master files will necessarily multiply into image files derived for specific tasks/purposes. Storing all this derived content in a pre-defined set of the corresponding folders from the very beginning and following an orderly naming convention will make it very easy to navigate the breadth of derived image data. The method outlined in Table 3 below allows for quick file sorting and filtering. Under this system each filename is meaningful and provides a sufficient amount of information to understand the nature of the file and its storage location. (See Table 3)
| File Type | Description | Naming Convention |
| Master | The retouched image file created from the RAW file. Usually TIFF or PSD. | RAWFILENAME_[V-versionNumber]_MASTER.TIFF or
RAWFILENAME_[V-versionNumber]_MASTER.PSD [V-versionNumber] can be optionally used to specify a particular Version of the Master file. If version number is used in the Master filename, it should be carried in the same format through all files derived from it, so the chain is never broken. Example: NU_JDE_A_001_MASTER.TIFF Or NU_JDE_A_001_V-01_MASTER.TIFF |
| Preview | A low resolution preview image of the raw file. JPG format. | RAWFILENAME_Preview.JPG Example: NU_JDE_A_Preview.JPG |
| Contact Sheet | JPG image or a multi-page PDF document consolidating preview images along with their file names and possibly additional image information. | Imaging_Type_ ModelNameAbbreviation_ PhotoshootSequence_ ContactSheet_ SequenceNumber Example: NU_JDE_A_ContactSheet_01.JPG or NU_JDE_A_ContactSheet.PDF |
Naming of all subsequent file types is based on the name of the Master file.
| Prints | Files created for prints of specific size. | MASTERFILENAME_PRINT_S-widthxheight-DPI-dpi_[COLOR/BW]_EL-EditionLimit
Example NU_JDE_A_V-01_PRINT_S-16x20_DPI-720_BW_EL-55 or NU_JDE_A_V-01_PRINT_S-16x20_DPI-720_COLOR_EL-55 |
| Web Image | MASTERFILENAME_WEB-[L,M,S, T]_[COLOR/BW] Where L,
M and S correspond to Large, medium, small and thumbnail size.
Another approach may be to use image width (or height) in pixels to indicate the size of the web image: MASTERFILENAME_WEB-800px_[COLOR/BW] Example NU_JDE_A_V-01_WEB-L_BW NU_JDE_A_V-01_WEB-800px_BW |
|
| Any Other Type | File names for other types of image files may be constructed following the simple schema described above. |
If a number of derived files for some RAW image becomes particularly large, you can create a special subfolder named after that RAW file, place it under the corresponding "4 - Finished works" folder, and move all corresponding Master and derived files there.
Hint: to enforce uniqueness of abbreviations used in naming the directories and files, one could follow a simple procedure. File system itself has a restriction on allowing only unique filenames in any single directory. Create a special directory, "Used_Unique_Abbreviations." Every time that you need to use a new abbreviation, create there a new bogus Text file (right-click, select "New/Text Document") with a desired abbreviation as its filename. The operating system will automatically verify this new abbreviation against the existing list of previously created files. If no errors occur and filename is allowed, the abbreviation is unique. Do not delete this new file as its filename becomes a part of the used abbreviations list.
Adapting to the described file naming scheme allows fast and easy location of folders containing any particular image. When names given to directories and files are meaningful, navigation through the digital content becomes straightforward. You would no longer need to search for files blindly, because you will simply know where they are.
As many of my Master files contain various adjustment layers and are stored in PSD format, the choice of an application capable of displaying previews of these and other derivative files correctly is pretty much limited to Adobe products, and Adobe Bridge, packaged along with Photoshop CS2 and CS3, can do the job. The filtering interface of Bridge has a lot left to be desired, but it does work.
An important configuration setting of Adobe Bridge is enabling usage of the distributed cache. This effectively instructs Bridge to put a pair of files (Adobe Bridge Cache.bc and Adobe Bridge Cache.bct) into each and every folder that it visited during searching. These files contain cached information specific to the image files stored in respective folders. This setting is quite beneficial since the cached files can travel along with the corresponding image files during backup, restore, and move operations and will stay available for immediate utilization by Adobe Bridge.
Once I had an unfortunate accident while updating Meta information for a Master image file using the Bridge application. The unexpected crash in the application caused corruption of the image file and rendered it unusable. Unfortunately, I attempted this before that particular file has been backed up. That was a hard lesson and since then I adhere to the policy to never update Meta information writable into the image file itself for any of the original or derived image files, with exception of easily re-creatable content, such as previews, contactsheets, and web images.
In my studio workflow, I use Adobe Bridge to get a quick and convenient overview for the body of work stored in various image formats. Reaping more benefits from the chosen file naming scheme, the filter criteria in my Bridge collections is primarily based on the meaningful parts of image filenames, such as abbreviation of the model's name, "MASTER", "PRINT", print sizes, "Preview", "WEB", etc.
There is a great variety of software packages available to handle the critical task of the Raw conversion. Primary differences between them are summarized below:
Upon comparing various raw conversion packages, I decided to use SilkyPix for the majority of the raw files. SilkyPix answered my demands for the three aspects listed above very well. There are rare exceptions when the magic of SilkyPix alone cannot rescue the shot (e.g. extraction of the most extreme dynamic range from the highlights of the raw data). In these cases, I turn to Adobe Camera Raw or try other solutions.
The power of SilkyPix lies in the high quality interpolation algorithm of the RAW data and user-selectable precision control (Demosaic Sharpness slider). A 100% setting will force extraction of the full resolution of data that seems possible. Because of heavy machinery employed in the raw processing algorithm of SilkyPix, conversion process is not a fast one. For a faster generation of the small JPG previews I use a 10% setting. However, when ultimate resolution of the Master file is desired, it is an indispensable tool.
SilkyPix offers non-destructive editing of image files in RAW and other image formats. Supported color space of input images is limited to sRGB and aRGB. Editing information for each file is stored in a subdirectory "SILKYPIX_DS," which gets automatically created in the folder where the edited file is located. (see Figure 6: SilkyPix RAW Converter Software)
The batch renaming feature in SilkyPix is very extensive and quick. When using batch renaming in SilkyPix all files in "SILKYPIX_DS" folder also get automatically renamed. However, renaming of images outside of SilkyPix may break the link between them and the files of their SilkyPix adjustments.
SilkyPix allows creating user presets for every type of image adjustment as well as user presets spanning multiple types of image adjustments; the latter ones are called "tastes." An example of a useful "taste" is a combination of Color Space, Sharpness, Tonal adjustments and white balance. Adjustments can be copied from one image to another with a simple copy-paste operation. Pasting can be partial so that only some of the copied adjustments would get pasted. This simple feature greatly speeds up RAW processing workflow.
Here is a quick walk through of my SilkyPix preparatory workflow for studio nudes. First, pick a representative RAW file of the photo session and apply the following rough adjustments:
Copy these adjusted settings (Ctrl-C). Select all raw files from the photo session and paste copied settings (Ctrl-V). Open each raw file and apply individual adjustments to exposure and any other settings necessary. Correct image rotation and cropping. Delete unnecessary images as you find them (you can also mark them for deletion by pressing F6 to re-view your selections later prior to mass deletion). When done, select all raw files and mark them for batch processing to generate Previews. Batch-develop all raw files into JPG format, sRGB Colorspace, in a size you prefer (I suggest to use resolution size of your monitor) with suffix "_Preview." When done, select the first raw file, set Demosaic sharp to 100% and output Colorspace to adobeRGB, and copy these settings (Ctrl-C). Select all raw files. Right-click and select "Paste partial development parameters." Check "Development (Demosaic sharp/Resolution Plus/Colorspace)." Press "Execute." Select menu "Option" (Alt-O) / Option (O) / Generate thumbnails of all scenes. This will generate all thumbnails to allow faster image browsing in SilkyPix. Now you are at the state when raw files are available for further final fine-tuning. Since cropping/rotation is already applied, comparison of various images for selection of the best one is simplified (JPG preview images can also be used for this purpose).
SilkyPix has a true 16-bit processing engine under the hood. The most popular professional photo editing application, Photoshop, surprisingly handles only up to 15 bits of the image data. Losing that 16th bit means that an entire HALF of the tonal information is ignored by Photoshop when the image file is opened and is subsequently discarded once it is saved. Considering the importance of the tonal information in creation of fine art images, I recommend using the full 16-bit power of the raw conversion process to create a full tonal base for the desired look of the Master image file while keeping an overall contrast of the image low and avoiding clipping by staying away from pure whites and blacks. The tonal and color adjustments in Photoshop should be applied only in adjustment layers.
Table 4 provides a quick overview of editing tasks that can be performed in RAW conversion process and in post-production manipulation of the master files.
| Procedure/Setting | Raw Conversion | Editing Master File |
| Recommended Tool | SilkyPix | Adobe Photoshop |
| Output Format | 16-bit TIFF, uncompressed | 16-bit TIFF, uncompressed or ZIP
compression PSD format if saved with adjustment layers. Optionally PSD files can be saved in maximum compatibility mode in which file also contains a flattened layer representing a merge of all layers in the file. This will increase the PSD file size. If adjustment layers do not contain masks, then masks can be deleted to minimize file size. |
| Exposure Correction | Apply as needed. Avoid clipping. | Fine-tune if needed. Global exposure variations in the image can also be corrected using curve adjustment layers applied in luminosity mode with gradient-filled masks. In difficult cases, raw conversion outputs made with various exposure settings can be merged with masking. Setting white and blank points (using a final curves adjustment layer) should be used only in a very last step of editing. Avoid clipping. Do not merge final curve adjustment layer with the background image layer. Image layer should remain at lower contrast, so it will be available for further manipulation if necessary. |
| White Balance | Fine tune as desired. | In difficult cases, raw conversion outputs made with various WB settings can be merged with masking (should not apply to studio setting). |
| Demosaic Sharp, Superresolution | 100%. This is equivalent to capture sharpening but
without any artifacts. 10% for generation of JPG previews. |
N/A |
| Resolution | Full | Full |
| Color Space | Widest possible. Adobe RGB. | Editing space: Prophoto RGB, Gray Gamma 2.2. Blend RGB colors using gamma 1.00 |
| Histogram | Always carefully observe. | Always carefully observe. |
| Rotation/Cropping | Apply in SilkyPix as it performs rotation without loss of resolution due to simultaneous enlargement of the file while rotating. | Fine-tune what was not fully handled in raw conversion. It is better to perform image rotation only once to avoid softening. |
| Lens Corrections | Correct chromatic aberrations, vignetting, minor distortion. | Correct anything that was not fully handled in raw conversion. |
| Color Mode | As desired. Do not over-saturate. | |
| BW Conversion | Do not use, it is better to have master file in color for any possible future use. | Apply using adjustment layers. Channel Mixer is generally sufficient for studio nudes as image colors are not pronounced. Keep color master as a background layer. |
| Selective Color Adjustment | As desired. Do not over-saturate. Avoid posterization. | Use Selective Color, Hue/Saturation adjustment layers in various blending modes. Use masks when needed. |
| Curves | As desired. Keep contrast lower than needed for the final output (prints, etc). Avoid clipping. | Apply in adjustment layers. Use masks if needed. |
| Gamma/Contrast | As desired. Keep contrast lower than needed for the final output (prints, etc). Avoid clipping. | Apply in adjustment layers (via levels or
curves). Use masks if needed. Color-based contrast enhancement can be applied using Hue/Saturation adjustment layer in luminosity blend mode. |
| Sharpness | All settings at 0. Do not apply in raw conversion.
In SilkyPix make sure to turn off Unsharp Mask filter in the final output (development) window. "Pure Detail" sharpening can be applied for JPG previews. |
Master Files: very minor sharpening to emphasize
selective content and compensate for texture loss due to
editing. Minor Unsharp Mask filter—apply in luminosity mode with large radius for local contrast enhancement; apply through masks in separate layer when needed. |
| Noise Reduction | As desired. Multiple conversions can be made with and without NR for further merging/masking. | As needed; use Noise Reduction with suppression of color and luminosity noise applied in separate layer in luminosity blend mode; use masks. |
| Moire (false color (FC) adjustment) | As desired. Multiple conversions can be made with and without FC for further merging/masking. | As needed; use Noise Reduction with suppression of color noise applied in separate layer in color blend mode; use masks. |
I find a pressure sensitive pen and a tablet to be an indispensable tool for fine retouching of digital images. They provide an incomparable control over the retouching process; no mouse can replace it. I am currenlty using a Wacom Intuos3 9x12" tablet and am extremely happy with it. I perform all editing of Master files in Adobe Photoshop, which provides a full support for these pressure sensitive tools.
Photoshop's Healing Brush is an excellent tool for an extensive localized retouching that studio-nudes images often require. It allows for touching up imperfections on model's skin while preserving the texture. The quality of texture preservation depends on the selected brush radius (in comparison to the characteristic size of the texture), the hardness setting of the brush, and the extent to which the brush is being used.
Dodge & Burn brushes are very instrumental for correcting unevenness of the skin (such as cellulite) without any loss of texture. Their use is also very effective for localized enhancement of dimensional appearance of a subject's form.
Brightness non-uniformities throughout the frame can be corrected using Curves adjustment layers with gradient-filled masks.
Below are a few illustrations of before and after comparisons using various techniques.
Example Image 1: Localized Exposure Correction
|
Example Image 2: Skin Retouching using the Healing Brush
|
Example Image 3: Using Healing Brush and Cloning Patches
|
Example Image 4: Using Healing Brush and Cloning Patches
|
There are numerous ways to perform a conversion of a color image to its black and white version because there are numerous ways to present variations of color and luminance in terms of variation of brightness in the grayscale. In my usual studio setup there is very limited color information in the photographic scene, hence, the conversion to black and white in most cases can be performed using a Channel Mixer adjustment layer. In some cases I also add either Hue-Saturation or a Selective Color adjustment layer in lightness blend mode for better transformation results. The final step in black and white conversion is Curves adjustment layer, which is used for fine-tuning the contrast and setting the black and white points of the image. It is important to observe the histogram of the image to avoid tonal clipping at all stages of the adjustments.
Although most of my sculptural nude works are printed in black and white, I do preserve the originally captured color information in the Master files, saved in this case in PSD format, along with all adjustment layers used for black and white conversion. When masks of the adjustment layers are entirely white, they can be deleted to decrease the size of the PSD file. Black and white adjustment layers can be grouped into a Layer Set and reused for conversion of similarly shot images from the same photo session—Photoshop allows drag-and-drop operation for layer sets between open image files.
Once Master files are created, retouched and adjusted, we can use them to derive images targeted for a specific output. The most common derivations are images that will get presented on the Internet and images that will be printed.
Creation of web images can be standardized to produce them in a desired size and framing preferences by employing Photoshop actions. A sample action for creating web images can be downloaded from my web site at the following link: www.michaelezra.com/downloads/actions/WebActions.atn
If you would like to create your own action, simply record it by going through all steps that you would like to be covered when creating a web image, and then play that action on a flattened Master file from which you would like to create a web image. (Figure 7: Creation of Action in Photoshop)
To create a Photoshop action, select Actions window (Alt-F9 for PC), click on a small triangle in the upper right corner as shown in figure below and select "New Action" menu; give a name to a new action. Action recording is started automatically, as indicated by a "Start" button (red circle) in the toolbar on the bottom of the action window. Execute all image processing steps that you would like to be recorded into this action and press "Stop" button (square) in action window toolbar when finished. The action is automatically saved when recording is stopped. To execute the action in order to repeat all recorded steps on another image, open the desired image and press the "Play" button (triangle) in the action window toolbar. Always test the new action on a test image first; specifically, be careful with "Save Image" steps in the action to avoid accidental overwriting of the original image.
A few notes concerning web images:
There are a lot of nuances and details that can be covered within this topic. In this overview article I will only point out a few essential aspects important for creation of high quality images targeted for printing.
I use 24-inch wide Epson Stylus 7800 printer with Epson K3 Ultrachrome inks for printing limited editions of the sculptural studio nudes. The quality of print output and its longevity in combination of these inks with archival inkjet paper makes this setup an excellent choice for producing high quality fine art prints.
Preparation of the print image is always specifically tuned to a particular printing setup. The factors to be considered include the printer, inks (if it is an inkjet printer), the paper, the driver settings, and, optionally, the lighting conditions in which prints will be viewed.
It is important to know the native resolution of the printer, which is going to be used. Epson printers, for example, support resolutions of 360 and 720 dpi. When an Epson driver is used for printing, it will interpolate the image to one of these resolution settings, unless the image is already prepared in either 360 or 720 dpi. Interpolation quality of the generic printer drivers is horrible and results in severe artifacts, clearly noticeable as jagged lines in the diagonal patterns of the image. This is why it is important to prepare a file in a specific size, matching the resolution expectation of the printer driver. The higher resolution settings usually mentioned in the driver (1440 and 2880 dpi) usually confuse the users as they have an entirely different meaning. These numbers refer to the accuracy at which each individual pixels' content gets rendered in mixing of various inks. Yet the pixels themselves are placed on paper at the printer's native "pixel" resolutions. Higher resolution setting in the driver will result in more precise rendering of each pixel and consequently—finer reproduction quality of tonal transitions in the prints.
There are many possible ways to obtain a quality interpolation of the images. Unfortunately there is simply no single method that I could recommend because image interpolation is very content specific. In my practice, I usually interpolate the same image using various methods and blend the result using Photoshop layers and masks. This allows me to maintain a high level of detail and texture where needed and smoothness in open areas of the image. Master image should be flattened prior to interpolation.
The final sharpening of the image should only take place when it has been interpolated to the final size. Smart Sharpen filter can provide good results but is extremely time consuming on large images. SilkyPix "pure details" sharpening method can be applied to TIFF images with excellent results and much faster than Smart Sharpen filter in Photoshop. I usually blend sharpened and not sharpened images together using masks to have an additional control on what gets sharpened and to what extent. (Example Image 6: Using Smart Sharpen Filter for Image Sharpening
The interpolation of the image to the target size and the following sharpening should be carried out while the image is in the editing color space (Prophoto, Adobe RGB or Gray Gamma 2.2).
It is important to understand that similar to the monitor & video card color profile that interprets image data for viewing, each printing setup has its own color profile that is used to interpret image data for printing.
The correct reproduction of the intended color depends on performing a correct interpretation of image data tuned specifically for the printing setup to be used. This interpretation is carried out through a conversion of the image to the output color profile. This color profile is created for each particular printing setup and is specific to driver settings, inks, paper and viewing lighting conditions being used.
For studio nudes I use the advanced black and white printing capability of the Epson driver. This allows me to print image data in Grayscale (Gray Gamma 2.2) mode instead of creating the custom color profiles for black and white printing. The puzzle of black and white printing gets solved by carefully selecting a finely-tuned combination of driver settings for media type, dpi setting and ink density for each inkjet paper used. This does require some experimenting, and in my case efforts were well rewarded with very fine and consistent quality print output. Advanced black and white printing of the Epson driver allows creation of an extremely archival output due to prioritized use of most archival black, light black and light-light black pigments.
I usually maintain a document for each of the images that I prepare for printing. These documents contain information on every significant setting used. This helps to ensure consistent printing of further editions of each of the images at a later time.
In addition to having a highly redundant main storage I would strongly recommend to have a few backups of the digital photographic content. Table 5 outlines the suggested approach for backup strategy.
| Backup Type | Schedule | Media/Notes |
| On-site | Tightly coupled with progression of workflow +
weekly Maintain backup log to help restore |
Magnetic - Attached External Storage (single hard
drive units or RAID 1) which is always turned off unless is used for
backup or restore. Optical - Archival quality DVD. |
| Off-site | Weekly, semi-monthly, monthly Maintain backup log to help restore |
Magnetic or optical - External Storage (single
hard drive units or RAID 1) transported off-site after
backup. Magnetic or optical - Off-site storage accessible via network. Magnetic or optical - Archival-quality DVD (use a brand different from one used for on-site storage. |
The digital photography workflow results in frequently occurring changes to the derived digital content: new master files get created and transformed into new print and web images all the time. From the storage point of view it means that the same organized directory structure is populated with more and more additional data. Hence, the most convenient form of backup for organized directory structure presented in this workflow would be an update backup. In this mode the backup target is maintained as a complete copy of the source, maintaining the same exact organized directory structure. This provides the advantage of a full understanding of the structure and content of the backups and to have a full control of the restore process as well, not relying on interpretation of the hidden backup structure forced by any specific backup software package.
In update backup mode the new files are copied over, changed files are overwritten with the latest changes and files that are deleted from the source (the main storage) are also deleted from the backup. The update backup is very fast due to the fact that unmodified files are skipped, thus the total amount of data being copied is limited to updates only.
My search for the most suitable software package capable of backing up large amounts of photographic data using update backup mode landed on a very lightweight freeware tool, called SameDir. Originally I invested in a Nero package, which claimed to be able to support update backup mode. It never really worked correctly in update mode and the performance of Nero's BackitUp application is on the slow side. The advantage of SameDir is that it is very fast and so far is reliable. SameDir also offers data compression options, which I decided not to use as compression significantly slows the process.
Nero Express, on the other hand, offers excellent tools for writing backups on DVDs. If you decide to compress files prior backing them up, I would recommend using a WinRar application. Backup of RAW files should be carried in a disk-at-once mode, as these backups are final. Backup of the derived content can be carried using the multi-session mode to allow adding of the files as they get created.
As backup DVDs get created, their sheer number can easily get overwhelming. It is important to keep them organized to be able to quickly locate the correct disk with a specific file. For this purpose I wrote a custom freeware application - Smart Disk Catalogue. It is a simple to use utility program, which allows creating a comprehensively searchable catalogue of backup DVDs. Smart Disk Catalogue can be downloaded from my web site at www.michaelezra.com.
For backing up the operating system drive I would recommend Acronis TrueImage Home. It supports a large variety of RAID controllers and if your system is installed on RAID 0, this is definitely the program of choice. Nero also offers system drive restoration, but that feature is primitive and is not compatible with RAID controllers.
Should a problem with the main storage occur, or some files on the main storage get accidentally overwritten, deleted or corrupted, a restore from backup is a sure help. Restoring from well organized backups is not a complicated process. Considering the fact that the directory structure of backups is the exact copy of the source in the main storage, the restore process becomes a straight-forward copy of the required folders and files.
Considering that on-site and off-site backups are carried on separate schedules, differences in their content may offer different versions of your photographic data. Versioning can also be accommodated with DVD-based backups, carried at various times, thus storing different versions of your data. Smart Disk Catalogue application can help you search through the content of your entire DVD library and find files with similar or the same names across all DVDs. The date and time stamp of the files listed in search results should allow you to locate the correct version easily.
I hope that the information provided in this article will be helpful to streamline your photographic process. The digital age brings the benefits of advanced technologies, efficient and detailed systematization and easier automation of the photographer's workflow. Careful and thorough analysis and adjustment of each and every step of your workflow will allow you to increase its efficiency, solve technical challenges, and free up your mind for creative artistic exploration of the photographic medium. After all, even though the technical side of the photographic process is fun by itself, what ultimately matters is the result that comes out of it!
Michael Ezra is a New York-based photographer specializing in fine art sculptural nudes as well as fine art landscape photography. Born in 1972 in Tbilisi, the capital of the former Soviet Republic of Georgia, he studied painting from his early years. He graduated Tbilisi State University with degrees in theoretical physics and painting, and continued to intertwine a scientific approach with an artistic expression ever since. In 1993, Michael Ezra moved to United States and began his physics studies in the field of medical imaging. For the next seven years, through his move to New York City and career change to software architecture and development, complex surrealistic painting remained his main channel of artistic self-expression. In 2000, inspired by Michelangelo sculptures, Michael Ezra turned to photography and began developing his concept of black-and-white studio sculptural nudes. In his easily recognizable style, he created a wide body of original sculptural works, which is sometimes referred to as a "reference catalogue of poses." In 2004, he extended his portfolio to include fine art landscape photography. Michael Ezra's professional background in imaging and in software development enabled him to make an easy transition to the world of digital photography and provided a solid foundation for comprehensive understanding of existing technologies. More »
Michael Ezra
|
Michael Ezra
|
Michael Ezra
|
Michael Ezra
|
Michael Ezra
|
Michael Ezra
|
Original text ©2008 Michael Ezra, screen shots, tables, and photography ©2008 by Michael Ezra.
Article created July 2008
Doug Settle 
, July 22, 2008; 10:02 A.M.
Michael Ezra 
, July 22, 2008; 10:36 A.M.
Doug Settle , July 23, 2008; 02:28 A.M.
Jorge Angel González Freyre , July 28, 2008; 01:00 P.M.
Michael Ezra , July 29, 2008; 01:47 P.M.
Debbie Silbert , July 30, 2008; 09:48 P.M.
frank menesdorfer , September 07, 2008; 01:01 P.M.
t ode , September 13, 2008; 01:54 P.M.
George Anchev 
, November 05, 2008; 10:33 A.M.
John Connelly , November 06, 2008; 10:46 A.M.
Terry McGovern , November 25, 2008; 10:35 A.M.
nathaniel mcmahon , December 01, 2008; 12:02 P.M.
Corey Rankin , December 11, 2008; 08:20 P.M.
