High-resolution color imaging presents high demands on lenses. One measure of the quality of a lens for a given application is its modulation transfer function. In color imaging, this will be different for different wavelengths of light. In a very basic sense, this is a measure of the sharpness of an image as affected by the lens.
On a recent project, I needed a simple way to compare the performance of three different lenses in an application that was to use a 3-CCD color linescan camera (JAI CV-L107CL). I devised a method by which a pattern, printed on a large piece of paper, was attached to a rotating drum. The drum was rotated in front of the camera and the sharpness of the image was determined by viewing a plot of the intensities of the red, green and blue channel signals, when imaging a fine set of black lines in the image. The set up is shown here.
The first lens that I tested was a 35 mm Nikon achromat. This is a lens that is designed to bring two wavelength of light (generally red and blue) simultaneously into focus. Furthermore, this lens is not really intended to be used in a 3-CCD linescan camera. The 3-CCD linescan camera has additional prism elements in its optical path for separating the incoming light into three separate paths (red, green, blue), each directed to a separate CCD. The resulting image of the black line target is shown here.
The next lens to be tested was a 35 mm Kowa apochromat. This lens is designed to bring three wavelengths of light (red, green and blue, in this case) simultaneously into focus. This lens also is designed to work with the prism used in the 3-CCD camera. The results were considerably better.
The third lens to be tested was a 35 mm BlueVision superachromat, which is designed to bring four wavelengths of light (near IR, red, green and blue in this case) simultaneously into focus. This lens, also, was designed for use in conjunction with a wavelength-separating prism. The results for visible light, in the current application, were essentially the same as for the apochromatic Kowa device and are shown here.
Since the additional capability of imaging infrared was not needed in this case, the considerably higher cost of the superachromat was not justified.