A new approach for image processing in foreign fiber detection

Authors:
Wenzhu Yang;Daoliang Li;Liang Zhu;Yuguo Kang;Futang Li
Affiliations:
College of Information and Electrical Engineering, China Agricultural University, Beijing 100083, PR China and College of Mathematics and Computer, Hebei University, Baoding 071002, PR China;College of Information and Electrical Engineering, China Agricultural University, Beijing 100083, PR China;College of Mathematics and Computer, Hebei University, Baoding 071002, PR China;China Cotton Machinery & Equipment Co., Ltd., China Cotton Industries Ltd., Beijing 100089, PR China;China Cotton Machinery & Equipment Co., Ltd., China Cotton Industries Ltd., Beijing 100089, PR China
Venue:
Computers and Electronics in Agriculture
Year:
2009

Citing 11
Cited 4

Evaluation and comparison of different segmentation algorithms

Pattern Recognition Letters
A segmentation algorithm for noisy images: design and evalution

Pattern Recognition Letters
Machine vision for automated visual inspection of cotton quality in textile industries using color isodiscrimination contour

Computers and Industrial Engineering
Fast image segmentation based on multi-resolution analysis and wavelets

Pattern Recognition Letters
Threshold selection using fuzzy set theory

Pattern Recognition Letters
Threshold selection based on cluster analysis

Pattern Recognition Letters
Image segmentation based on merging of sub-optimal segmentations

Pattern Recognition Letters
Design and application of industrial machine vision systems

Robotics and Computer-Integrated Manufacturing
An adaptive image segmentation algorithm for X-ray quarantine inspection of selected fruits

Computers and Electronics in Agriculture
Image segmentation evaluation: A survey of unsupervised methods

Computer Vision and Image Understanding
Automated vision system for localizing structural defects in textile fabrics

Pattern Recognition Letters

An improved genetic algorithm for optimal feature subset selection from multi-character feature set

Expert Systems with Applications: An International Journal
Original paper: Classification of foreign fibers in cotton lint using machine vision and multi-class support vector machine

Computers and Electronics in Agriculture
Fast recognition of foreign fibers in cotton lint using machine vision

Mathematical and Computer Modelling: An International Journal
An image segmentation method for apple sorting and grading using support vector machine and Otsu's method

Computers and Electronics in Agriculture

Quantified Score

Hi-index	0.00

Visualization

Abstract

In the textile industry, different types of foreign fibers may be mixed in cotton that need to be sorted out to ensure the quality of the final cotton textile products. Automated visual inspection (AVI) system is a popular tool at present for real time foreign fibers detection in lint. The image processing is one of the key techniques in the AVI system. This paper presents a new approach for processing images of foreign fibers. This approach includes four main steps, namely, image transformation, image enhancement, image segmentation and segmentation post-processing. In the first step, color images were captured and then transformed into gray-scale images. In the second step, the histograms of the gray-scale images were analyzed, and a piecewise nonlinear transform model was proposed to enhance the image based on the analysis results. Thirdly, an improved Otsu's method was employed for segmenting the gray images of foreign fibers. Finally, post-processing of segmentation was performed, and a modified closing operation in mathematics morphology was proposed to fill up the image gaps of the wirelike foreign fibers caused by segmentation, and an area threshold method was suggested to remove the small objects generated by pseudo-foreign-fibers. The results indicate that the contrast between objects and background can be remarkably improved by the proposed enhancement model. The Otsu's method can effectively subdivide the enhanced images into objects and background. The gaps can be filled up by the modified closing operation, and the small objects can be removed effectively using the Area Threshold algorithm.