A Dense Litchi Target Recognition Algorithm for Large Scenes

Figure 1

YOLOv4 network output. The parameters contained in each bounding box are as follows: the center coordinates of the box (t_x, t_y), the width , and height (t_h) of the target, the confidence score (P₀) that the box has a target, and the probability (P_c) that the target in the box belongs to each type of object.