The first step to identify the sick chicken in the farms by machine vision system is segment of object from images fast and correctly. However, it is a challenge to extract the chicken from pictures because of the complex background. A segmentation method based on the difference of three components of RGB model and HSV model was presented to extract yellow feather broilers from the image. Totally 200 images were taken under the natural environment by using digital cameras and iPhone 6. Totally 30 images were selected from 200 images to setup pixels data sets for the color components analysis. Background and feather data sets included 10 sample areas in each selected image. Each sample area had 10×10 pixels. Comb data sets had three sample areas of each selected image and included 5×5 pixels for each sample area. All data sets were analyzed in the different color models, such as RGB, HSV, L*a*b*. It was found that the value of R, G, B components of the background and the chicken belly was nearly the same or very close while the average value was different. This characteristic was used to abandon the background pixels in the RGB model. Then the remaining part of the image was converted to the HSV color model. The research obtained H component threshold for comb and feather by statistics data sets, respectively. Totally 102 images were processed in the experiment. The result showed that segmentation accuracy of yellow feather broilers from images using the proposed method was 86.3%, which was better than that of L*a*b* color model (78.4%). This method was simple with short calculation time and was suitable for real-time segmentation.