Paddy-brown rice boundary detection method based on machine vision

Abstract

Objective: To detect and locate the boundary line between paddy and brown rice in separation equipment. Methods: A method based on straight-line edge detection was proposed by performing model matching with different color features. Converted the resulting images to grayscale and applied the linear edge detection algorithm to locate the target position. Results: Our approach reduced the error to within 5% and significantly improved recognition accuracy compared to traditional methods. Conclusion: Proposed method has a small recognition error, making it suitable for identifying the boundary between paddy and brown rice.

Keywords

• boundary
• paddy separation
• edge detection
• machine vision