Research Article

[Retracted] Fast Recognition Method for Multiple Apple Targets in Complex Occlusion Environment Based on Improved YOLOv5

Figure 10

Results of identification and positioning of apples under different lighting angles by four algorithms. (a) Recognition effect of HOG + SVM algorithm under different illumination angles. (b) Recognition effect of Faster RCNN algorithm under different illumination angles. (c) Recognition effect of YOLOv6 algorithm under different illumination angles. (d) Recognition effect of the baseline YOLOv5 algorithm under different illumination angles. (e) Recognition effect of the improved YOLOv5 algorithm under different illumination angles.
(a)
(b)
(c)
(d)
(e)