Detecting Temperature Changes in Liquor Distilling Process
Author Information
Author(s): LIU Xiaolian, Gong Shaopeng, Hua Xiangxu, Chen Taotao, Zhao Chunjiang
Primary Institution: Taiyuan University of Science and Technology
Hypothesis
Can a lightweight model improve the detection of temperature changes in the surface layer of fermented grains during liquor distillation?
Conclusion
The YOLO v5ns-DP model effectively improves the accuracy and speed of temperature detection in the distillation process while reducing computational resource usage.
Supporting Evidence
- The YOLO v5ns-DP model improved mAP by 0.6 compared to the original YOLO v5n model.
- The model achieved a detection speed of 89.891 FPS.
- The number of parameters, GFLOPs, and model size were reduced by 28.6%, 16.5%, and 26.4%, respectively.
Takeaway
This study created a smart way to check the temperature of grains used in making liquor, helping to make better liquor faster.
Methodology
The study used infrared imaging and a YOLO v5n model, enhanced through knowledge distillation and channel pruning, to detect temperature changes in fermented grains.
Limitations
The model still faces challenges with false detections and leakage issues that need further optimization.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website