Improving the network architecture of YOLOv7 to achieve real-time grading of canola based on kernel health
Angshuman Thakuria, Chyngyz Erkinbaev
Abstract
The occurrence of heated and immature canola kernels caused by excessive drying and frost damage is undesired by grain buyers due to low oil yield and diminished market value. The current grading process is visually examining each kernel's endosperm colour and counting the damaged seeds. As this process is time-consuming, the current study proposes an automated grading technique based on multi-object detection, tracking, and counting. The detection task was achieved via an improved YOLOv7 network (YOLOv7_ours) to increase its performance in accurately identifying small objects and decrease its computational cost and size; by adding two convolutional block attention modules and substituting convolutional layers with ghost layers in all the Efficient Layer Aggregation Networks modules, and in the Spatial Pyramid Pooling Cross Stage Partial module present in YOLOv7. The detection weights were fed to the ByteTrack multiple object tracker to track the detections frame by frame in a video feed. The unique identities generated by the tracker for each detected object of interest were then used to count the number of defects using a line cross algorithm. The mean average precision ([email protected]) obtained after training the YOLOv7_ours model was 1.02% better and its cost and size were 32.1% and 37.1% lower than the baseline YOLOv7 model. In a test video, the tracking model achieved a multi-object tracking accuracy of 84.8% and the counting accuracy was determined to be 93.9%. This three-stage model can be readily deployed in an edge device for accurate and real-time grading of canola kernels by grain buyers.