Improved Pest Detection Method for Low-Quality Cotton Images
Author Information
Author(s): Ma Yukun, Wei Yajun, Ma Minsheng, Ning Zhilong, Qiao Minghui, Awada Uchechukwu
Primary Institution: Henan Institute of Science and Technology
Hypothesis
Can the DCP-YOLOv7x method improve pest detection accuracy in low-light environments?
Conclusion
The DCP-YOLOv7x model significantly enhances the detection precision of cotton pests in low-light conditions.
Supporting Evidence
- The DCP-YOLOv7x model achieved a detection precision of 95.9% for cotton pests.
- Mean Average Precision (mAP@0.5) was improved to 95.4% under low-light conditions.
- Experimental results showed significant improvements over the baseline YOLOv7x model.
Takeaway
This study created a new method to help farmers find pests on cotton plants even when it's dark, making it easier to protect their crops.
Methodology
The study used a combination of image denoising and enhancement techniques followed by a modified YOLOv7x object detection model.
Limitations
The model may struggle in highly complex environments with significant occlusions.
Statistical Information
P-Value
p<0.05
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website