An Adaptive Spherical Image Transform for High-Speed Fruit Defect Detection

A novel adaptive spherical transform was developed and applied in a machine-vision apple defect sorting system.The image transformation compensates the reflectance intensity gradient on curved objects and provides flexibility in copingwith fruits natural variations in brightness and size. Test results demonstrate the effectiveness of the proposed method forhigh-speed defect detection on apples. This transform can be applied to other fruits or similar curved objects.