The apparent sharpness of a micro image created from a document is of critical importance for microfilm imaging and retrieval systems. Fuzzy characters and blurred images make end users wish they had the original hard copy. Image resolution is the first characteristic that the end user will notice and is thus worthy of our best effort. Here we will focus, no pun intended, on the three factors that affect image resolution: film, camera and processing.
All three silver microfilm manufacturers (Agfa, Fuji & Kodak) certify their medium speed microfilms to have the ability to achieve 800 lines/mm of resolution. High-speed films generally can achieve 600 lines/mm and some silver duplicating films can achieve 1,000 lines/mm. It is important to remember that these numbers are produced in a laboratory environment and seldom relate exactly to user results. In fact resolution is more likely to be used to describe the entire imagining system; the concept of acutance is used to describe the film’s characteristics.
We will discuss resolution measurement later but first it is good to understand acutance. When describing image edge characteristics, the term 'acutance' should be used. Acutance refers to the ability of a film to reproduce an edge with as abrupt a transition between the background and the line as possible.
Acutance is a property that is normally measured by the film manufacturer. The special targets and micro-densitometers, which are required for measurement, make acutance overly complex for the end user. Acutance, unlike subjective visual readings, provides objective data on a particular film's imaging qualities. 'Objective' in the sense that acutance data is machine read and thus human variability is eliminated. Presently, modulation transfer function, abbreviated MTF, is the most popular method for measuring image edge characteristics. Unlike a knife edge test which evaluates a film's ability to reproduce one edge, MTF utilizes multiple edges.
Testing of modulation transfer function requires the use of a very high quality target that contains a sinusoidal wave pattern. Higher spatial frequencies place greater demands on the ability of a film to reproduce line information (good correlation with micro imagery). The sinusoidal target is brought into close contact with the film to be tested and is exposed to a point light source. After the test film has been processed, it is scanned with a micro-densitometer (similar conceptually to a conventional densitometer except that the reading aperture is very small -- less than 10 microns). The micro-densitometer provides an objective evaluation of the test films ability to copy the target. If the film performance were ideal, it would copy the original target with no loss in wave amplitude for the higher frequencies. Since no film is perfect, there is some loss in wave amplitude for the higher frequencies. Modulation transfer function is a measurement of the film ability to copy (transfer) this high frequency wave pattern. MTF data is available for all silver films.
It is important for the silver film user to appreciate the interrelationship between image acutance, photo speed and contrast. If image characteristics were the sole goal, silver films would be manufactured with very thin emulsions containing very small grains. This would result in a film with high acutance, but this film would also have slow photographic speed and high contrast. Silver film manufacturers must balance image characteristics against the need for film speed and reasonable contrast (especially important for microfilming photographs or poor quality documents). Recognize your trade off; if you require very high acutance, your film will generally be slow and higher in contrast. Fast films will be lower in acutance and lower in contrast.
Recording devices come in all sorts of sizes and shapes and can be divided into two general categories: static and dynamic. In static cameras neither the subject to be filmed nor the film move during exposure. Planetary cameras are in this category. Exposure in dynamic cameras takes place while the subject and the film are moving. This system works satisfactorily provided the subject movement and the film movement are synchronized. Rotary cameras are in this category.
Static cameras resolve finer details than dynamic cameras. The performance of dynamic cameras is affected by vibration, lack of planarity between the film plane and subject plane and less precise positioning of the film in relation to the optics.
The film and the subject in a static camera are positioned so that during exposure they are perfectly parallel and stationary. If either is slightly out of alignment, resolution loss at the edge of the image will result. This reduces vibration and eliminates synchronization problems. Many static cameras contain special vacuum platens that curve the film plane slightly. This curvature of the film is designed to overcome limitations in lens design and provide better resolution edge to edge. Within a camera class, either static or dynamic, various techniques can be employed to improve imaging performance. The results your camera provides are a product of both design features and proper operation. The design features should be evaluated during the camera selection process while proper operation is a function of good process control and maintenance.
Camera Factors that Affect Resolution
Clean Optics and Mirrors
Thumb prints; dust and moisture deposits on the front element of the camera lens will individually or collectively degrade the lens ability to capture fine details. Mirrors are in all rotary cameras and a few planetary cameras. Lens and mirror cleaning should be part of a regular maintenance program.
Cameras should be isolated from vibrations. Even minor vibrations can cause a loss in image resolution. Take for example, a planetary camera used to film engineering drawings, located on the fourth floor of a 19th century mill. The camera produced quality output, except that on occasion poorly resolved images appeared. After a careful investigation, it was discovered that the degraded images were related to vibrations caused by a forklift truck working on the third floor. The solution was not very sophisticated; filming was rescheduled to times when the forklift was inoperative. This solution was suitable for a part-time facility but obviously unacceptable for a continuous operation. In most instances vibrations are cured either through relocation of equipment or through systems which dampen the vibrations before they reach the camera.
Cameras that allow a range of reduction ratios usually have an automatic focusing mechanism. As the camera head is moved up and down its column, the lens is automatically refocused for the change in the distance between object and film plane. If the resolution is acceptable at one reduction ratio but badly degraded at a different ratio, the auto focus mechanism may be out of order. Also the lens in non-automatic cameras can be knocked out of focus from rough handling. If you’re in doubt that the camera is working to full potential have a qualified technician run a bench test.
Depth of Field
Each lens, depending on its focal length and aperture, has a particular depth of field -- the range of distances from the camera within which the subject will remain in sharp focus. For some cameras this zone might extend no more than 2" above the copy board. If you were to film a thick bound book, it could easily extend beyond the 2" depth of field zone causing poor focus.
Incorrect Film Loading Procedures
When loading film into the camera make sure that the film is properly seated in its guide channels. Also be careful not to knock the pressure plate assembly causing either misalignment or disruption of the vacuum system if the camera is fitted with such a device. Finally, make sure that the emulsion is facing the lens and that the film channels are adjusted for the film size.
Image density, assuming consistent development, should be controlled through exposure. Resolution is adversely affected by excessively high density, e.g., > 1.50, because the higher densities tend to accentuate the granularity of the image. Densities lower than .80 lack the necessary image contrast for good readability.
While the influence of the film and camera on image resolution is easily recognized, many fail to appreciate the importance of development. In fact, both the processing conditions and the type of developer will affect the image resolution.
Underdevelopment tends to promote smaller grain size. Both reduced dwell time and/or lower development temperature can be used. This technique, however, does have its limitations because reduced development also lowers contrast. Though an underdeveloped image may in fact contain smaller silver grains, the human observer may not be able to notice any improvement because of reduced contrast. A higher contrast image is always perceived as 'sharper' by a casual observer. If you intend to affect grain size by changing development conditions, I would recommend trying a lower temperature, i.e., 10°F, and observing the results. Some film speed may be lost; therefore an adjustment for exposure may become necessary. For most users, the point of this section is don't overdevelop the film because grain size will increase and resolution will be affected adversely.
Most microfilm developers are designed to provide good film speed but also high contrast. Active developers produce photo speed by causing the individual grains to clump together into larger density sites. This clumping improves speed but affects resolution adversely. There are special developers, fine grain developers, which minimize clumping and contain silver solvents that physically reduce grain size. The trade-off when using these developers is a loss in photo speed and lower contrast. For most users, special fine grain developers are only of academic interest. Microfilm developers are formulated to give the best balance between resolution, film speed and contrast and it is seldom the case that the combination film/developer is the limiting factor for image resolution.
Resolution is defined as the ability of a photographic system to record fine details. The system resolution can be no better than the weakest link usually the optical system in the camera. Image resolution is measured by filming a standard target (the NBS target continues to be the most popular), processing the film and evaluating the results with a microscope. The target contains a series of horizontal and vertical lines that are equally spaced. The largest pattern, 1.0, contains I line and space pair per millimeter, while the smallest pattern, 10.0, contains 10 line and space pairs per millimeter. The series is actually an assembly of ISO Resolution test charts spaced through out the target to test the entire lens.
After filming, the target is read with a microscope that has a magnification twice the reduction ratio of the image. For example, a 24X target should be read with a 50X scope. Target evaluation requires the observer to pick the smallest line pattern in which both vertical and horizontal lines are separated. The most consistent method for resolution reading is to scan the line patterns looking for complete line fill-in, either horizontally or vertically (ideally they degrade at the same rate). Once the first pattern to fill-in is detected, drop back to the next larger pattern. Note the number next to the pattern and use that number to calculate lines per millimeter. Since the target has been reduced in size, the number next to the pattern no longer provides actual lines per millimeter on the film. For example, the 5.0 pattern on the life-size target indicates 5 line and space pairs per millimeter. After filming, the chosen target must be multiplied by the reduction ratio to determine lines per millimeter. If the 5.0 pattern is selected and the target has been reduced 24 times, the correct reading would be 120 lines per millimeter. Now that we understand how to determine system resolution some important observations are in order.
Resolution readings provide data on the capabilities of your imaging system. It is important to remember this point: resolution data is not merely a property of the film. Resolution can also be employed as a process control tool. For proper process control, a target should be included with every job and the resolution values should be read and recorded. Absolute numbers are less important for process control than the consistency of the reading from test to test. Consistency is a proof of the smooth functioning of cameras and film processors.
Regarding acceptable resolution limits: two factors must be taken into consideration. What resolution capability is necessary to capture the data; and what resolution is a particular system capable of generating. Obtaining required resolution is dependent on the quality of the original document, font size and reduction ratio. A good clean document imaged at 24X may need only 60 lines per millimeter for readability, while a bank check imaged at 40X may need 100 lines per millimeter for legibility. (More discussion on resolution can be found in ANSI/AIIM MS-23).
Finally, resolution readings are valuable process control tools, but it must be recognized that resolution characterizes only one part of the imaging system. Proper density control is equally important. Many users spend an inordinate amount of time worrying about resolution when for most applications 100 lines/mm, are adequate. Remember that films with very high resolution can also be high in contrast and lack adequate photo speed. The best technique for film selection is to run practical tests and pick the film that has the best information transfer. If you look solely at resolution, you may not have the best film for your application.