DIGITAL IMAGE ANALYSIS BASED ON AUTOMATED COUNTING CLUSTERED SOYBEAN SEEDS

Abstract

-  This paper investigates the use of digital image analysis techniques for developing for counting clustered soya bean seeds. Images are extracted from source WWW.21food.com). As manual counting have several issues such as low accuracy and higher cost. Automated counting techniques give a fast and low cost of counting soya bean seeds. This paper follows 5 processing steps. First process converting the image into gray scale and thresholding is applied using CLAHE. Second dilation is applied to enhance the image. Third applied masking to enhance the image. Fourth edge detection algorithm is applied. Fifth step beans extracted with respect to bean shape.