Related Experiment Video
Updated: Jul 8, 2025
11:02
Fruit Volatile Analysis Using an Electronic Nose
Published on: March 30, 2012
21.5K
Geographical traceability of soybean: An electronic nose coupled with an effective deep learning method
Huaxin Sun1, Zhijie Hua1, Chongbo Yin2
1School of Automation Engineering, Northeast Electric Power University, Jilin 132012, China; Bionic Sensing and Pattern Recognition Team, Northeast Electric Power University, Jilin 132012, China.
Food Chemistry
|December 17, 2023
Summary
This study introduces an adaptive convolutional kernel channel attention network (AKCA-Net) with an electronic nose (e-nose) for soybean quality traceability. This method accurately identifies soybean quality based on origin, preventing fraud and ensuring quality standards.
Area of Science:
- Agricultural Science
- Artificial Intelligence
- Sensory Science
Background:
- Soybean quality is linked to geographical origin, leading to market fraud.
- Accurate traceability of soybean quality is crucial for agricultural integrity.
Purpose of the Study:
- To develop an effective system for soybean quality traceability.
- To leverage advanced AI for identifying soybean origin-based quality differences.
Main Methods:
- Utilized an electronic nose (e-nose) to collect gas profiles of soybeans from various origins.
- Developed an adaptive convolutional kernel channel attention (AKCA) module for feature extraction.
- Proposed the AKCA-Net, integrating the AKCA module for deep gas channel interdependency modeling.
Main Results:
- AKCA-Net achieved high accuracy (98.21%), precision (98.57%), and recall (98.60%) in soybean quality recognition.
- Demonstrated superior performance compared to existing attention mechanisms.
- Successfully identified soybean quality based on origin-specific gas information.
Conclusions:
- The combination of AKCA-Net and e-nose offers a robust and accurate solution for soybean quality traceability.
- This approach effectively combats the issue of substituting low-quality soybeans.
- The proposed method ensures authenticity and quality standards in the soybean supply chain.
