A NEW METHOD FOR BRUISE DETECTION OF APPLES BY HYPERSPECTRAL IMAGING
Bruising is a severe problem that happens during harvesting and transporting of fruit. Visible and near-infrared reflectance spectra were used to detect the bruises on ‘Fuji apples. In the previous study, it was found that the relative reflectance of the samples between 738 and 812nm had fine linearity. By comparing the slopes of regression lines of the bruised to the non-bruised surface of apples, it was discovered that the slopes of the bruised are significantly higher than that of the non-bruised apples. Based on the previous study, the slope difference (SD) between bruised and non-bruised surface were studied. The results show that elapsed time after bruising, more than impact energy and fruit temperature, offered a potential to increase the SD within 7 days. The bruises of apples can be detected after 30min by the SD method. The SD showed more correlation with color difference ΔE*ab (R2=0.6063) than bruise volume (R2=0.2804).
Apple Bruise Detection Slope of Regression Line Hyperspectral Imaging
Xuan LUO Shuhuai ZHANG Ming SUN Teruo TAKAHASHI
The United Graduate School of Agriculture Science,Iwate University,Hirosaki,036-8561,Japan Faculty of Agriculture and Life Science,Hirosaki University,Hirosaki-shi,036-8561,Japan College of Information and Electrical Engineering,China Agricultural University,PO Box 63,17 Tsinghu
国际会议
北京
英文
1-6
2009-10-14（万方平台首次上网日期，不代表论文的发表时间）