Color and texture for corn seed classification by machine vision

Machine vision has been applied to various food materials inspection process of agricultural industry in order to achieve fast and accurate operation. In this paper, we proposed a method to classify more than ten categories of seed defects by using color, texture features and support vector machine (SVM) type classifier. We also developed an image capturing machine that is able to support large volume of seed samples. The image capturing machine was designed to control uncertainty light level, reflection and shadow appeared on seed samples. Therefore, quality of images that can be obtained from the machine in term of accurate color and exposure is high. In addition, the designed image capturing machine also provides support to background subtraction and touching object segmentation processes. In the image classification part, color histograms in RGB and HSV color space together with texture based on Grey level co-occurrence matrix (GLCM) and Local binary pattern (LBP) is adopted as features. The proposed systems were evaluated from more than 10,000 sample images. The obtained accuracies are 95.6% for normal seed type and 80.6% for group of defect seed types. The preliminary results of this study are useful information for future development of the quality control technique in practical usage.