Contrastive classification
A method for image recognition and automatic classification in computer science is called contrastive classification . The contrast of the image is increased step by step and the blurred image (because it has lost information) is compared with known patterns. In contrast to other classification methods (such as cluster analysis ), however, the aim is only to classify the entire object on the image without dividing it into smaller subclasses (global classification).
algorithm
Contrast increase
The process of increasing the contrast can be determined mathematically in various ways. The linear approximation of the three basic colors (RGB) is most frequently used. A divisor (d) is defined with which the number of steps can be influenced. If the color of the pixel is described by (r, g, b), then the color value results in the next step according to the formula . And analogously for g and b.
classification
- The resulting pattern is compared with the known patterns.
- If there is no match, a new run of the contrast increase is started.
- As soon as an identical pattern is found, the target class is found.
- If no identical pattern is found up to the maximum contrast, the generated patterns are checked backwards for known similar classes.
- As soon as the remaining areas are equal, the target class has been found.