Fine-grained non-motor vehicle feature detection method, storage medium and computer equipment

The invention belongs to the field of intelligent transportation, and relates to a fine-grained non-motor vehicle feature detection method, a storage medium and computer equipment, and the method comprises the steps: constructing a feature detection network which comprises a convolutional neural network, a plurality of block fusion modules and a driver and non-motor vehicle feature detector, wherein the block fusion modules which are connected are used for achieving a feature pyramid network from the deep feature map to the shallow feature map; selecting a plurality of network feature layers from the feature detection network, performing layer-by-layer blocking and fusion processing to obtain a driver fine-grained feature map and a non-motor vehicle fine-grained feature map, respectively inputting the driver fine-grained feature map and the non-motor vehicle fine-grained feature map into a driver feature detector and a non-motor vehicle feature detector for positioning and classifying regions of interest, and respectively outputting the regions of interest to an upper sampling layer; and further partitioning and fusing features to the shallow feature map to obtain a driver feature map and a non-motor vehicle feature map which are fused with finer-grained resolution information. According to the method, a detection network with block feature pyramid fusion is provided, and the target detection accuracy is improved.