No abstract

Thinning is an important agronomic process in pear production, thus the detection of pear inflorescence is an important technology for intelligentization of blossom thinning. In this paper, images of buds and flowers were collected under different natural conditions for model training, and the images were augmented by data augmentation methods. Model training was performed based on the YOLOv5s network with coordinate attention mechanism added to the backbone network and compared with the native YOLOv5s, YOLOv3, SSD 300, and Faster-RCNN algorithms. The mAP, F1 score and recall of the algorithm reached 93.32%, 91.10%, and 91.99%. The model size only took up 14.1 MB, and the average detection time was 27 ms, which are suitable for application in actual intelligent blossom thinning equipment.

To solve the problem of mechanized weeding in trunk type pear orchard, combined with the technology of stubble elimination and hydraulic obstacle avoidance, the authors designed an obstacle avoidance mower assembly. This paper carried out a field performance evaluation test to verify its usefulness. Six indexes were tested to evaluate its working performance. The results were: inter-row crushing rate of 89.99%, intra-row miss cutting rate of 2.42%, stubble stability coefficient variation of 4.25%, working efficiency of 0.32 hm2/h, fuel consumption of 16.25 L/hm2, profitable area of 0.75 hm2. The study could provide a reference for orchard mechanized weeding.