The behavior of the IS (r) selection index through the interval 1 ≤ r < 255. For small values of r the interval of tolerance of similarity ST(r) does not include the necessary tonalities that the image identification procedure requires. Therefore identification of pixels within an image can be expected to be imprecise. Consequently reduced values of λ
(r), will be expected, which lead to large values of the IS(r) selection index. For r ≥ 25, values of IS(r) decrease until its minimum value is attained at r = 128. According to the DCM selection criterion for the present data, both an RGB color format and ST(128) interval of tolerance of similarity can be used for efficient estimation of areas of eelgrass leaves using images with noise induced by humidity contents.