Abstract
Aiming at the problems of low accuracy of strawberry fruit picking and large rate of mispicking or missed picking, YOLOv5 combined with dark channel enhancement is proposed. In “Fengxiang” strawberry, the criterion of “bad fruit” is added to the conventional three criteria of ripeness, near-ripeness, and immaturity, because some of the bad fruits are close to the color of ripe fruits, but the fruits are small and dry. The training accuracy of the four kinds of strawberries with different ripeness is above 85%, and the testing accuracy is above 90%. Then, to meet the demand of all-day picking and address the problem of low illumination of images collected at night, an enhancement algorithm is proposed to enhance the images, which are recognized. We compare the actual detection results of the five enhancement algorithms, i.e., histogram equalization, Laplace transform, gamma transform, logarithmic variation, and dark channel enhancement processing under the different numbers of fruits, periods, and video tests. The results show that combined with dark channel enhancement, YOLOv5 has the highest recognition rate. Finally, the experimental results demonstrate that YOLOv5 is better than SSD, DSSD, and EfficientDet in terms of recognition accuracy, and the correct rate can reach more than 90%. Meanwhile, the method has good robustness in complex environments such as partial occlusion and multiple fruits.
Highlights
Strawberry [1] is a general term for strawberry plants of Rosaceae
Tomeet meetthe therequirements requirementsof of all-day all-day picking, picking, low-illumination low-illumination enhancement enhancement of of strawstrawTo berry images collected at night is required
The dark channel enhancement and histogram equalization enhancement basically restore the visual effect under illumination conditions and effectively change the gray value of each region, which meets the basic requirements of human vision, but analyzing the evaluation indexes of both, the dark channel enhancement effect is better than the histogram equalization enhancement; the Laplace enhancement effect is slightly worse, which is reflected in its enhancement of local shadows, and the Laplace method
Summary
Strawberry [1] is a general term for strawberry plants of Rosaceae. Strawberry fruit is conical with bright red color, while flesh color is light red with medium hardness. Using the HIS color space model [3], Zhao Ling extracted the color histogram of strawberry images’ H component, combined it with BP neural network, and established the strawberry maturity detection model based on color. Shortcomings lie in the fact that the detection model established by using HIS color space requires high illumination, and the effective recognition of strawberry ripeness cannot be under condition of low illumination. TheHIS algorithm in this paper lie completed in the fact that thethe detection model established by using color space requires addresses the shortcomings of the existing technology by considering and improving the high illumination, and the effective recognition of strawberry ripeness cannot be comfollowing three fruit category perfection, band and environpleted under theaspects: condition of low illumination. Illumination to enhance the accuracy of strawberry recognition at night
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
