Improving the Detection and Positioning of Camouflaged Objects in YOLOv8

Feature (linguistics) Convolution (computer science)
DOI: 10.3390/electronics12204213 Publication Date: 2023-10-12T07:14:32Z
ABSTRACT
Camouflaged objects can be perfectly hidden in the surrounding environment by designing their texture and color. Existing object detection models have high false-negative rates inaccurate localization for camouflaged objects. To resolve this, we improved YOLOv8 algorithm based on feature enhancement. In extraction stage, an edge enhancement module was built to enhance feature. fusion multiple asymmetric convolution branches were introduced obtain larger receptive fields achieve multi-scale fusion. post-processing existing non-maximum suppression address issue of missed caused overlapping boxes. Additionally, a shape-enhanced data augmentation method designed model’s shape perception Experimental evaluations carried out datasets, including COD CAMO, which are publicly accessible. The exhibits enhancements performance 8.3% 9.1%, respectively, compared model.
SUPPLEMENTAL MATERIAL
Coming soon ....
REFERENCES (35)
CITATIONS (6)
EXTERNAL LINKS
PlumX Metrics
RECOMMENDATIONS
FAIR ASSESSMENT
Coming soon ....
JUPYTER LAB
Coming soon ....