Automatic separation of defective eggs from qualified eggs reduces visual control difficulties which are done using human power as well as ensures improvement on the quality control process. In this paper, eggs are classified as dirt eggs, crack eggs and clean eggs as well as separation of defective eggs from qualified eggs. Hybrid solution is applied for detection defects on eggshell. Firstly, egg is separated from ground by using color image processing techniques. Then, dirt on dirty egg is separated with color recognition and then is said dirty egg by defined a threshold value. If there isn't dirt on the image, separated egg from the ground has entered the second control within the program and Sobel edge detection algorithm by applying eggs have been found in parts of the edges and cracks. Then again given a threshold value is decided whether it is clean egg or crack egg.