A New High‐Speed Foreign Fiber Detection System with Machine Vision

Abstract

A new high-speed foreign fiber detection system with machine vision is proposed for removing foreign fibers from raw cotton using optimal hardware components and appropriate algorithms designing. Starting from a specialized lens of 3-charged couple device (CCD) camera, the system applied digital signal processor (DSP) and field-programmable gate array (FPGA) on image acquisition and processing illuminated by ultraviolet light, so as to identify transparent objects such as polyethylene and polypropylene fabric from cotton tuft flow by virtue of the fluorescent effect, until all foreign fibers that have been blown away safely by compressed air quality can be achieved. An image segmentation algorithm based on fast wavelet transform is proposed to identify block-like foreign fibers, and an improved canny detector is also developed to segment wire-like foreign fibers from raw cotton. The procedure naturally provides color image segmentation method with region growing algorithm for better adaptability. Experiments on a variety of images show that the proposed algorithms can effectively segment foreign fibers from test images under various circumstances.

Contaminant and foreign fiber detection in cotton using Gaussian mixture model

“…Ref. [6] gives a nice overview of a system starting from sensors and illumination, up to blowing out of foreign materials. A short review can also be found in [7].…”

Detection Method of Smart Meter Carrier Module Pin Based on Blob Analysis

“…Stroppa et al used pyramid template matching algorithm to identify pins in electrical connectors and added multi-template strategy to deal with changes in pin types, shapes, colors, and surface smoothness [2]; Guo et al used template matching algorithm to identify the pins in electrical connectors and called different template matching methods according to different connector types [3]; Zhao et al used a combination of deep neural networks and template matching strategies to achieve fast positioning of pins in electrical connectors [4]; Zhang et al used k-means clustering to cluster the pins in hyperspectral images and detected problematic pins by comparing them with standard templates [5]; Fan Liangyu et al identified the centroid coordinates of pins in connectors using Blob algorithm and compared them with standard templates to detect the regularity of pins [6]; Xu Peng et al addressed the problem that threshold segmentation cannot be applied due to the differences in gray scale values of pins by using dynamic threshold segmentation and then selected features to locate the pin area. Finally, they obtained the center coordinates of the pin area by fitting a circle [7]; Li Huipeng et al performed subpixel edge detection of the bright spots of the pins using Zezike moments and used the center of the ellipse fitted by the edge points as the position feature point of the pin [8]; Zhang Yuan et al used laser triangulation to obtain point clouds of connectors, clustered the point clouds according to the number of pins, and finally fitted the coplanarity plane by RANSAC algorithm to the centroid of each cluster. The qualification of the product was judged based on the distance between various points and the fitted plane [9]; Fang Jianlong et al used Canny algorithm to detect the outer contour of the connector terminal and filtered out interference points by bubble sort to reduce system error [10].…”

GPU Accelerated Intuitionistic Fuzzy and Otsu Algorithms for Foreign Leaf Detection in Cotton

“…However, this system was not quite efficient due to the differences between the chemical and physical characteristics of wool and cotton. The two problems of quality and speed, encountered in other previous studies, are an important distinguishing factor (Wang, 2015), (Chen, 2010). In this study, using GPU technology which can run hundreds of times faster than CPU were tried to solve this speed problem (NVIDIA, 2019).…”