Estimation of mango crop yield using image analysis – Segmentation method

This paper presents an approach to count mango fruit from daytime images of individual trees for the purpose of a machine vision based estimation of mango crop yield. Images of mango trees were acquired over a three day period, 3 weeks before commercial harvest occurred. The fruit load of each of fifteen trees was manually counted, and these trees were imaged on four sides. Correlation between tree counts and manual image counts was strong (R2 = 0.91 for two sides). A further 555 trees were imaged on one side only. For these images, pixels were segmented into fruit and background pixels using colour segmentation in the RGB and YCbCr colour ranges and a texture segmentation based on adjacent pixel variability. Resultant blobs were counted to obtain a per image mango count. Across a set of 555 images (with mean ± standard deviation of fruit per tree of 32.3 ± 14.3), a linear regression, (y = 0.582x − 0.20, R2 = 0.74, bias adjusted root mean square error of prediction = 7.7) was achieved on the machine vision count relative to the image count. The algorithm decreased in effectiveness as the number of fruit on the tree increased, and when imaging conditions involved direct sunlight. Approaches to reduce the impact of fruit load and lighting conditions are discussed.

Graphical abstractFigure optionsDownload full-size imageDownload as PowerPoint slideHighlights► Estimating fruit load on mango trees from images. ► Algorithm based on colour and texture features. ► Tested across 555 images, machine vision count compared with manual photo count. ► R2 of 0.74 and RMSEP of 7.7 achieved (y = 0.582x − 0.20). ► Algorithm effectiveness decreased at high fruit loads, and under strong sunlight.