MEASUREMENT OF PIG WEIGHT BASED ON IMAGE ANALYSIS
Measurement of pigs weight is very important in pig production. But the traditional method, driving the pigs to a balance, requires much labor and, most importantly, annoys the pigs, which in turn affects their growth. To avoid this problem, this paper presents a new method that utilizes the technologies of image processing and computer vision. Firstly, the projected areas of the pigs image captured directly from the above are computed and the corresponding pigs weight is measured by a real scale. Secondly, the relation between the projected area and the weight are derived from the linear regression. Finally, for a new pig image, its projected area is computed and inputted to the linear regression equation that outputs the estimated weight. The results show that a strong relationship (with correlation coefficient 0.94) exists between pigs weight and the projected area after removing the sections of head and tail in images. By comparing with the real weight, the relative error is less than 2.8％. The experiment indicates that this hands-off method has great significance in scientific management of the pigs production which does not require large labor and material resources, and also avoid stressing the pigs.
pig weight measurement computer vision image processing projected area
Yan Yang Guanghui Teng Baoming Li
China Agricultural University, Beijing, 100083, China
国际会议
北京
英文
366-370
2005-10-14（万方平台首次上网日期，不代表论文的发表时间）