Abstract: Machine vision have been successful applied to monitor crop morphological and physiological status, such as leaf area index, nitrogen content, chlorophyll content and so on. The trials was conducted from 2004 to 2006 at the experiment station of Shehezi University located in Shihezi, Xingjiang Province to monitor cotton canopy chlorophyll information under field conditions by processing cotton population digital image, which is one measurement method with advantages of convenient, real time, quick and nondestructive. In order to obtain uniform cotton canopy digital images, the assistant device was used. In the field, the cotton canopy images were taken by a digital photo camera (OLYMPUS) at the squaring stage, early flowering stage, full flowering stage, peak boll stage and opening boll stage, respectively. The color characteristics of cotton population images were extracted with the image processing software developed by our lab. The correlation between color parameters of cotton canopy digital image and chlorophyll content of cotton functional leaf was analyzed. The results showed that the correlation of the color characteristics such as G－R, (G－R)/(G+R), r/g, g/r, and g－r in the RGB color system, and Hue in the HIS color system with chlorophyll content of functional leaf was significant at P<0.01. There also was a significant correlation between the population greenness index (PGI) and color parameter. The chlorophyll predicted models were established. The tested results about the regression models suggested that G－R was the best parameter to monitor cotton population chlorophyll information. The relative error of chlorophyll content and PGI estimations was about 6.96% and 11.60%, and RMSE was 0.1138 and 0.1643. The chlorophyll content predicted model was y=－1.3008+0.2125(G－R)－0.0038(G－R)²（R²=0.8669**）, and PGI predict model was y=－0.9726+0.1227(G－R)－0.0016(G－R)²（R²=0.7487**）.

Key words: Cotton population, Machine vision, Chlorophyll content, RGB, HIS, Percent ground cover of image