The Method of the Calculating of Frequency Characteristics of Image Gaining and Processing Systems

Abstract

The basic parameter for deflning the quality of the image is its resolution which can be under the in∞uence of: † conditions in which the image was gain (the conditions of exposure); † parameters of image gaining system; † parameters of processing system. Quantitative estimation of the mentioned in∞uences is essential in order to estimate the quality of the gaining system and image processing. That operation is possible by calculating the modu- lation transfer function (MTF) of the whole system. The calculation is provided via the research of the suitability of the impulse response of the system to the model signal. The model signal for the image is a suitably prepared test, which is the subject for observation and then for electronic processing into digital form. Peak resolution of a digital image is deflned by the parameters of the applied optical and detection system in the flrst place, and then the parameters of the processing devices such as scanners, converters and recorders. The gaining of strip tests images is the classic method of researches on in∞uences of these devices' parameters and conditions of acquiring on the image parameters. However, in case of remote sensing imageries it is required to have done the tests of properly large sizes and covered with colours of deflnite contrasts. This is expensive and long-lasting procedure, but most of all subjective. The quality assessment of the gained image is carried out through the optical observation conducted by a certain group of people and then calculating the average of observation results. In the article the authors suggest applying a new method for this kind of researches, based on deflning modulation transfer function (MTF) of the gaining and image processing system on the basis of test images of much smaller sizes and easier to execute. The image is feather edge between the areas of maximum contrast. The principle advantage of the method is the objectivity of measurement, which guarantees its repeatability.