Printer Friendly

Real time process monitoring and fault detection using linear array sensor with FPGA for textile industries.


Textile industries use only pure cotton for their fabric production. Cotton while it is harvested has many types of contaminants in it. The quality of the fabric is affected even if the color of the cotton is contaminated.

In olden days, the color contamination in the cotton was identified by the labor's working in the industry. Due to technological advancement the detection method is also automated. Many vision systems have been proposed for the advancement of the detection method.

One of the novel approach for detecting the cotton contaminants uses X-ray micro-tomographic system. A computer vision algorithm is used to classify the contaminants using high resolution. it is used in offline mode and the main shortcoming of this method is that it is time consuming.

Multi-wavelength Imaging System (MIS) is used for the detection of the contamination in the cotton in the spectral region. Each type of contaminant has different absorption and reflection characteristics. In this method an image fusion algorithm based on wavelet analysis is adopted. Through experimental research a desirable range of wavelength was determined.

Texture feature based detection method was also proposed. In this method the camera acquires the image of the cotton and it is then given to an image processing system. A Gray Level Co-occurrence Matrix algorithm (GLCM) is used to detect the sharp contrast objects in the acquired image. A rotating search filter is implemented to remove the unwanted edges and to locate the impurities in the image. This method has a drawback of computational load.

One of the detection method used color spaces for detecting the contaminants in the cotton. A recognition algorithm is used to detect the impurities and the RGB images obtained are converted Y CbCr color space. In To Cite This Article: S. Sajitha and Dr. M. Jayasheela., Real Time Process Monitoring and Fault Detection Using Linear Array Sensor with FPGA for Textile Industries. Advances in Natural and Applied Sciences. 10(10); Pages: 297-301 this y represents gray image, Cb and Cr represents blue and red color components respectively. The green color component Cg is constructed from the available components. For the removal of noise a shape feature algorithm is adopted and it also identifies the false points in the image. The shortcoming of this method is that white colored contamination cannot be identified.

The advancement of the above detection method is that instead of YCbCr color space HSI color space is used for detection of white colored contaminants present in the cotton. HSI denotes Hue Saturation and Intensity of the image. A progress in this method could be made by designing am system which differentiates the contamination based on the material.

Gaussion mixture model uses the color and intensity of the obtained image as a distribution. This model is not suitable when the distribution of color and intensity has multiple nodes in it.

Another detection method proposed was based on image processing technique. In the contamination recognition technique three color spaces are used. HSI, YCbCr and gray image. The detection method follows 3 steps picture transformation followed by image processing based on wavelet transform and post processing of the image. Wavelet transform is used to format the color of the image and to adjust the image histogram. In the method real time automation is required.

In this paper we present a detection method based on a sensor array to make the detection method efficient and to thereby to increase the production.

The remainder of this paper is organized as follows: section 2 briefly discusses the proposed system hardware architecture. In section 3 the simulation results are proposed and in section 4 the conclusions are given.

Proposed System:

2.1 Block Diagram of the system:


Figure 1 shows the proposed system block diagram. In order to detect the contamination in the cotton it is placed over the conveyor. LSA receives the acquired image from the camera. The sensor scans the image of the cotton and identifies whether the cotton is contaminated or not. The sensor and microcontroller are interfaced to each other. The analog output from the sensor is sent to the controller which converts it to a digital value and is displayed in the LCD. If the cotton is contaminated then the microcontroller sends a command signal to the pneumatic valve which blows off the contaminated cotton into the trash outlet.

2.2 Hardware Architecture of the Proposed System:

The figure 2 shows theHardware Architecture of the proposed Vision System.From the figure it is clear that the proposed vision system consists of cameras with lighting system arranged on both sides of the cotton conveyor, a pneumatic valve and the supporting electrical and mechanical interface for the inspection machine.


The cotton continuously moves over the conveyor. LSA receives the acquired image from the camera. The light source provides the necessary illumination. Linear sensor array scans the color of the cotton. The LSA are available in a variety of lengths and pixel resolutions (DPI). For comparing white/black color thresholds the analog output is interfaced directly to an ADC (Analog to Digital) converter.

The scanning is done based on index value. The color variations are identified by the sensor based on the index values. An index value below 40 indicates that the cotton is contaminated.

The ADC in microcontroller converts the analog output from the LSA into a digital value and is displayed in the LCD. At regular time interval the LCD is updated by the current status from microcontroller.

In case of any contamination the unwanted cotton is blowed off from the conveyor with the help of the pneumatic valve, which is commended by the microcontroller.


This section deals with simulation and results of the proposed method. . The result shows that the index value based detection method using LSA is efficient. The schematic is designed using PROTEUS7.1 which is a virtual modeling system. The necessary codingfor the schematic is programmed in AVR STUDIO 5.0 and is linked to the schematic for simulation results.


The Figure 3 shows the Circuit diagram of the proposed detection method. The schematic diagram of the circuit is designed using PROTEUS VSM 7.0 software. Itcan be inferred that the microcontroller is connected to the sensor. The necessary ground, clock values are given. The sensor is connected to the ADC (analog to digital conversion) pin. The sensor gives analog output to the controller which converts the analog index value to a digital value and sends to the LCD. The value in the LCD makes clear whether the cotton is pure or not. If suppose the cotton is not pure then it is removed by blowing pressurized air through pneumatic valve.


The Figure 4 shows the simulation output of contaminated cotton detection. From the Figure 4 it is clear that when the sensor detects a contamination its value will be less than 13 in VOUT and the LED glows. The sensor value is sent to the controller. The controller converts the analog value to a digital value and it displays the status of the cotton as BLACK in the LCD so that the contaminated cotton would be removed by pneumatic valve.


The Figure 5 shows the simulation output of contaminated cotton detection. From the Figure 5 it is clear that when there is no contamination the value of the sensor would be greater than 13 in VOUT and the LED is off. The sensor value is sent to the controller. The controller converts the analog value to a digital value and it displays the status of the cotton as WHITE in the LCD.

Conclusion And Future Scope:

In this paper work real time process monitoring and fault detection using a Linear Sensor Array is proposed to categorize the pure and contaminated. The sensor detects the contamination in the cotton based on the threshold values. On comparison with other systems the proposed approach is easy and requires less processing time. The LCD helps in displaying the status of the cotton. The manual removal of cotton is replaced by pressurized air from pneumatic valve.

The future scope of this system is that an FPGA can be replaced in place of microcontroller that scans simultaneously and thereby increases production


[1.] Aditi Sachar and Sugandha Arora, 2012. A review of automatic cotton contaminant detection techniques, International Journal of Computer Science and Information Technology & Security, 2(4): 384-387.

[2.] Chengliang Zhang, Xianying Feng, Lei Li and Yaqing Song, 2010. Identification of Cotton Contaminants Using Neighbourhood Gradient Based on YCbCr Color Space, 2nd International Conference on Signal Processing Systems (ICSPS), 3: 733-738.

[3.] GajabeVaishali Vasant and Manoj. D. Patil, 2015. cotton contaminants automatic identification techniques,.InternationalJournal of Emerging Technology and Innovative Engineering, 1(5): 421-425.

[4.] Gao, W., Z.H. Wang, X.P Zhao and F.M. Sun, 2008. Robust and efficient cotton contamination detection method based on HSI color space, Research gate, 7(30: 729-735.

[5.] Guo, J.X. and Y.B. Ying, 2008. 'Progress on Detecting Technique and Sorter of Raw Cotton Foreign Matters', Transactions of the Chinese Society for Agricultural Machinery, 7(4): 107-113.

[6.] Huang Jing and Jiang Ming, 2014. A research on detection method of cotton contamination based on improved RBF neural Network, SixthInternationalConference on Measuring Technology and Mechatronics Automation, 1: 256-261.

[7.] Jia, D.Y. and Ding, 2005. Detection of foreign materials in cotton using a multi-wavelength imaging method', Measurement Science and Technology, 6(3): 1355-1362.

[8.] Mingxiao Ding, Wei Huang, Bing Li, Shaohong Wu, Zhiqiang Wei and Yunkuan Wang, 2009. An Automated Cotton Contamination Detection System Based on Co-occurrence Matrix Contrast Information, IEEETransaction Circuits System I: Regular Papers, 54(1): 13-25.

[9.] Pai, A., Sari-Sarraf, E.F. Hequet, 2004. 'Recognition of cotton contamination via X-ray micro-tomographic image analysis, IEEETransactions on Industry Applications, 1(5): 77-85.

[10.] Pooja Mehta and Naresh Kumar, 2011. 'Detection Of Foreign Fibers And Cotton Contaminants By using Intensity And Hue Properties', International Journal of Advances in Electronics Engineering, 1(1): 230240.

[11.] Strolz, H.M., 2000. 'ITMF cotton contamination survey 1999', Proceeding of Cotton Conference in Texas USA. Los Angeles, USA: World Textile Publication Ltd, pp: 63-67.

[12.] Tantaswadi, P., J. Vilainatre, N. Tamaree and P. Viraivan, 1999. Machine vision for automated visual inspection of cotton quality in textile industry using color isodiscrimination contour', Computers and Industrial Engineering, 12: 347-350.

[13.] Zivkovic and Zoran, 2004. 'Improved adaptive Gaussian mixture model for background subtraction', IEEE Proceedings of the 17th International Conference on Pattern Recognition, 2(6): 665-669.

(1) S. Sajitha and (2) Dr. M. Jayasheela

(1) Department of Information and Communication Engineering, KIT- Kalaignar Karunanidhi Institute of Technology, Coimbatore-641 402

(2) Department of Electronics and Communication Engineering, KIT- Kalaignar Karunanidhi Institute of Technology, Coimbatore-641 402

Received 27 May 2016; Accepted 28 June 2016; Available 12 July 2016

Address For Correspondence:

S. Sajitha, Department of Information and Communication Engineering, KIT- Kalaignar Karunanidhi Institute of Technology, Coimbatore-641 402

COPYRIGHT 2016 American-Eurasian Network for Scientific Information
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2016 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Sajitha, S.; Jayasheela, M.
Publication:Advances in Natural and Applied Sciences
Date:Jun 30, 2016
Previous Article:Clustering based dynamic duty-cycled multiple-rendezvous multichannel MAC (DMM-MAC) for Bursty traffic in underwater sensor network.
Next Article:Trust among militants in wireless sensor network.

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters |