Related Experiment Video
Updated: Jul 2, 2026
07:17
Exploring the Application of Surface-enhanced Raman Scattering-based Biosensing of Individual sEVs in Disease Diagnosis and Therapeutics
Published on: March 13, 2026
[A method of recognizing biology surface spectrum using cascade-connection artificial neural nets].
Wei-Jie Shi1, Yong Yao, Tie-Qiang Zhang
1Laser Information Technology Research Center, Shenzhen Graduate School, Harbin Institute of Technology, Shenzhen 518055, China. steven2wn@yahoo.com.cn
Guang Pu Xue Yu Guang Pu Fen Xi = Guang Pu
|August 30, 2008
Summary
This study introduces a novel artificial neural network system for identifying apple pericarp spots using visible light spectra. The system achieves over 85% accuracy, even with noise, improving upon traditional methods.
Area of Science:
- Agricultural Science
- Computer Science
- Spectroscopy
Background:
- Accurate identification of fruit defects is crucial for quality control in agriculture.
- Traditional methods for analyzing fruit surface spectra often suffer from low accuracy and poor noise resistance.
Purpose of the Study:
- To develop an advanced spectrum recognition system for detecting defects on apple pericarps.
- To improve the accuracy and anti-noise capabilities in spectral analysis of biological surfaces.
Main Methods:
- Utilized a fiber-probe spectrometer to capture visible spectra (500-730 nm) of apple pericarp spots.
- Developed a three-level cascade-connection artificial neural network (ANN) system for spectrum recognition.
- Employed fuzzy mathematics for an objective expression of recognition results.
Main Results:
- Successfully recognized spectra corresponding to rotten, scarred, and bumped spots on apple pericarps.
- Achieved a recognition accuracy exceeding 85%, even under a 15% noise level.
- The cascade-connection ANN system demonstrated superior performance compared to single ANN systems.
Conclusions:
- The proposed cascade-connection ANN system offers a robust and accurate method for spectral analysis of fruit surface defects.
- The novel expression method based on fuzzy mathematics provides objective and precise recognition outcomes.
- This technology has the potential to enhance automated quality assessment in the fruit industry.
