A new method of boundary tracking in fruit detection.
Deng JiZhong, Zhang TaiLing
Author Affiliation: College of Polytechnic, South China Agricultural University, Guangzhou 510642, China.
Journal of South China Agricultural University 22 : 80-82
Abstract : A modified boundary tracking method for fruit detection was proposed. In this method, a sample in an operation image that could either be a binary image or a grey image was partitioned into four regions. To accelerate boundary tracking, the priority of tracking directions in each region was determined by the appearance of the sample. Tests using images of fruits, such as apples, pears and mangoes, were performed. The results confirmed that the time of processing a 150×150 image of pear was less than 1 second.