Related Experiment Video
Updated: Sep 11, 2025
11:02
Fruit Volatile Analysis Using an Electronic Nose
Published on: March 30, 2012
21.8K
An Integrated Lightweight Neural Network Design and FPGA-Accelerated Edge Computing for Chili Pepper Variety and
Ziyu Guo1, Yong Yin2, Haolin Gu1
1College of Artificial Intelligence, Southwest University, Chongqing 400715, China.
Foods (Basel, Switzerland)
|August 14, 2025
Summary
A novel system using an electronic nose and FPGA accelerates chili pepper variety and origin detection. This intelligent agricultural solution offers high accuracy and efficiency for market applications.
Area of Science:
- Agricultural Science
- Computer Engineering
- Sensory Science
Background:
- Chili pepper markets face challenges with variety confusion and origin ambiguity.
- Accurate identification is crucial for quality control and consumer trust.
Purpose of the Study:
- To develop an integrated system for rapid and accurate chili pepper variety and origin detection.
- To address market issues through intelligent agricultural technology.
Main Methods:
- Utilized an electronic nose (e-nose) to collect gas data from diverse chili pepper samples.
- Developed a lightweight convolutional neural network (CNN) model, ChiliPCNN, for classification.
- Integrated the ChiliPCNN model with a field-programmable gate array (FPGA) for hardware acceleration.
Main Results:
- ChiliPCNN achieved high accuracy rates: 94.62% for variety identification and 93.41% for origin tracing.
- Optimized FPGA implementation reduced latency to 5600 ns and power consumption to 1.755 W.
- Demonstrated superior classification performance and stability compared to other deep learning methods.
Conclusions:
- The proposed FPGA-integrated e-nose system provides an effective solution for chili pepper authentication.
- This technology advances intelligent agricultural management, promoting automation and efficiency in the sector.
