Target detection method, electronic device, and automatic retail system
By introducing a coordinate attention mechanism and a path aggregation network structure into the object detection algorithm, the accuracy and efficiency of small object detection are improved, solving the problem of balancing detection accuracy and efficiency in large-format visual detection scenarios.
CN122416094APending Publication Date: 2026-07-17HISENSE RONSHEN (GUANGDONG) FREEZER CO LTD
View PDF 0 Cites 0 Cited by
Patent Information
- Application Number
- CN202610406501.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-03-30
- Publication Date
- 2026-07-17
Smart Images
Figure CN122416094A_ABST
Abstract
The application provides a target detection method, an electronic device and an automatic retail system. The method comprises: acquiring a scene image of a target scene; inputting the scene image into a trained target detection model; using a plurality of feature extraction modules to perform multi-scale feature extraction on the scene image to obtain a plurality of basic feature maps of different scales; using a neck network to perform bottom-up and top-down bidirectional feature fusion on the plurality of basic feature maps to obtain fusion feature maps of different scales; using a plurality of detection heads to respectively perform target classification and positioning prediction based on the fusion feature maps of different scales to obtain respective prediction results; and determining the category information and image position information of each target object in the scene image according to the prediction results of the plurality of detection heads.
Need to check novelty before this filing date? Find Prior Art