Convolutional neural networks have recently experienced successful development in the field of computer vision. In precision agriculture, apple picking robots use computer vision methods to detect apples in orchards. However, existing object detection algorithms often face problems such as leaf shading, complex illumination environments, and small, dense recognition targets, resulting in low apple detection rates and inaccurate localization. In view of these problems, we designed an apple detection model based on lightweight YOLOv4—called Improved YOLOv4—from the perspective of industrial application. First, to improve the detection accuracy while reducing the amount of computation, the GhostNet feature extraction network with a Coordinate Attention module is implemented in YOLOv4, and depth-wise separable convolution is introduced to reconstruct the neck and YOLO head structures. Then, a Coordinate Attention module is added to the feature pyramid network (FPN) structure in order to enhance the feature extraction ability for medium and small targets. In the last 15% of epochs in training, the mosaic data augmentation strategy is turned off in order to further improve the detection performance. Finally, a long-range target screening strategy is proposed for standardized dense planting apple orchards with dwarf rootstock, removing apples in non-target rows and improving detection performance and recognition speed. On the constructed apple data set, compared with YOLOv4, the mAP of Improved YOLOv4 was increased by 3.45% (to 95.72%). The weight size of Improved YOLOv4 is only 37.9 MB, 15.53% of that of YOLOv4, and the detection speed is improved by 5.7 FPS. Two detection methods of similar size—YOLOX-s and EfficientNetB0-YOLOv3—were compared with Improved YOLOv4. Improved YOLOv4 outperformed these two algorithms by 1.82% and 2.33% mAP, respectively, on the total test set and performed optimally under all illumination conditions. The presented results indicate that Improved YOLOv4 has excellent detection accuracy and good robustness, and the proposed long-range target screening strategy has an important reference value for solving the problem of accurate and rapid identification of various fruits in standard orchards.

In order to improve the obstacle avoidance and endurance capability of quadrotor UAVs performing tasks such as forest inspection and rescue search, this paper proposes improvements to address the problems of too many traversed nodes, too many redundant corners, too-large turning angles and unsmooth generated paths in the traditional A* algorithm in path planning. The traditional A* algorithm is improved by using a segmented evaluation function with dynamic heuristic and weighting processing, a steering cost heuristic function based on heading angle deviation control, a redundant turning points removal strategy, and a quasi-uniform cubic b-spline. Through the test comparison of different complexity map scenarios, it is found that the improved A* algorithm reduces the number of traversed nodes by 64.87% on average, the total turning angle by 54.53% on average, the path search time by 49.64% on average, and the path length by 12.52% on average compared to the traditional A* algorithm, and there is no obvious turning point in the path. The real-world applicability of the improved A* algorithm is confirmed by comparing the effect of different algorithms on obstacle avoidance in a map of a real plantation forest environment.