1.1 Background of Fabric inspection systems The textile industry is a most important and rapidly growingsector that makes a considerable impact to the economical state of Sri Lanka.Quality is a major parameter in textile, thus good quality products increasethe profit of the industry as well as the customer satisfaction. If the defectsin the fabric are not detected properly significant financial loses can beoccurred. Hence, the fabric inspection process should carry out vigilantly. Thekey issue, therefore, is how and under what circumstances fabric inspectionwill lead to quality improvement. As a result, the automated fabric inspectionsystems have been designed to improve the efficiency of the inspection process.
Until, recently,the fabric inspection is still undertaken manually by skilled staff with amaximum accuracy of only 60%-70%. The modern fabric manufacturing industryfaces a lot of challenges due to high productivity as well as high qualitymanufacturing environment. Because the production speeds are faster than everand because of the increase in roll sizes, manufacturers must be capable ofidentifying defects, locating the sources of defects, and making the necessarycorrections in less time so as to lessen the amount of low quality fabric. Thisin turn places a greater strain on the inspection departments of themanufacturers. Due to the factors such as tiredness, boredom and carelessness,the staff performance is often unreliable. Therefore, the best reliableevaluation is through the application of an automated inspection system.
From the earlybeginning, the target is to achieve optimum potential benefits such as highquality, low cost, comfort, accuracy and speed in the manufacturing process. Asthe technology is revolutionized the fabric inspection process has beendeveloped from manual to automated machinery to help achieve all those benefitsin the manufacturing process. The application of automated fabric inspectionwould seem to offer a number of advantages, such as improved quality, reducedlabor costs, the elimination of human errors and increase the profit of theindustry. Therefore, the automated visual inspection is gaining progressiveimportance in fabric manufacturing industry.An automatedinspection system usually consists of a computer-based vision system. Becausethey are computer-based, these systems do not undergo the drawbacks of humanvisual inspection.
The application of digital image processing is useful intextile manufacturing and inspection. In recent years, it has proven to be themost promising, rapid and reliable solution for the development of automatedfabric inspection systems. Considerable efforts have been taken to developand/or improve the task of automated fabric inspection systems.
As all fabrichas the periodic regular structure, analyzing the parameters such as variance,intensity of the fabric presents a possible way to predetermine the occurrenceof defects in the fabric.The describedmethod in this thesis represents an effective and accurate approach toautomatic defect detection. It is capable of identifying various types ofdefects by monitoring the fabric structure. Presence of a defect over theperiodical structure of a fabric causes changes in the variance and intensityparameters of the fabric. By thresholding the defected area to create a binaryimage, it is possible to identify the exact location of the defect.
Further,for accurate detection of the defects the noise removal, morphological erosionand dilation, connected component analysis are carried out. The fabric defectcould be simply defined as a change in or on the fabric construction. Fabric defects can occur due tomachine faults, color bleeding, yarn problems, scratch, poor finishing, dirtspots, excessive stretching, and crack points. Because of the wide variety ofdefects, it will be useful to apply the study on the most major fabric defects.The chosen major fabric defects are: hole, missed-yarn, oil mark, knittingfault and pin holes.A fabric fault detection system designed with MATLAB isused for this procedure. It is implemented on the above mentioned chosen fabricdefects as well as the defect-free samples to identify as defect or non-defect.To verify the success of the defects it is implemented on one hundred and fiftysamples.
1.2 Fabric inspection Fabric inspection or defect detection is a process ofidentifying and locating defects in fabric. A fabric defect is a result of themanufacturing process. The textile industry is very concerned with quality.
Itis desirable to produce the highest quality goods in the shortest period oftime possible. Fabric quality inspection process is very important formanufacturing industry.Mainly, fabric defect detection uses two type of inspectionmodel. The first one is the Human based Inspection Systems (HIS).The secondsystem is Automated based Inspection Systems (AIS). 1.2.1 Human based inspection systems(HIS) In modern textile industry, fabrics are manufactured incomplex forms with high quality and high production line.
Thus, the inspectionprocess becomes more difficult and complicated. Therefore, industrial fabricinspection process has extremely high requirements. Figure 1.1 Manual fabric inspection system Traditionally, Human basedinspection systems are based on human.
This process is performed bywell-trained human inspectors. Usually, after the produced fabric is doffedfrom the weaving machine, it is batched into large rolls and sent to theinspection department. Most industries have power driven fabric inspectionmachines where the manufactured fabric rolls are unrolled on an inspectiontable(under adequate light) at a relatively higher speed of 8-20 meters perminute9. When a defect is found on the moving fabric, the inspector stops themachine to record the defect and its location, and starts the motor again. Foreach inspected fabric roll, the number of defects per meter length iscalculated and the fabric is classified.
Extra ordinary defect rate orrepeating defects are informed to the production department so that appropriateactions can be taken to reduce the defect rate. 1.2.2 Drawbacks of human based inspection systems While,modern fabric industries demonstrate a high productivity and high qualitymanufacturing environment, human based inspection systems have no ability to satisfytoday’s requirements due to limitations based on the physiological nature ofhuman. Therefore, Human based inspection systems (HIS) suffers from manydrawbacks described as follows:1) Trainingphase takes much long time to be a skilled inspector.
2) Thecontinue inspection becomes difficult and inefficient due to limitations inhuman behavior such as boredom and tiresomeness.3) Humanperceiving speed is slower when compared to machines hence the process takeslonger time.4) Thehuman based inspection system could never reach 100% of defect detectionaccuracy.5) Theinspector sometimes not capable of identifying the level of defects that isacceptable.6) Thelow fabric inspection is not compatible with high speed production line.
7) Humaninspectors have to deal with extensive variety of defects.8) Humanbased inspection systems are limited in finding defects and hence cause toreduce the profit of the fabrics. As aresult of these drawbacks several attempts are made to replace the manualinspection systems by automated systems to increase the accuracy and efficiencyof the process by increasing the quality level of fabrics. 1.2.3 Automated fabric inspection systems The shortcomings such as low profit, low accuracy and veryslow performance of manual fabric inspection process have led to the necessityof automated inspection systems. Many researches are carried out in recentyears to develop automated inspection systems with high accuracy, stability andspeed with respect to manual inspection systems.
Beside this, automated inspectionsystems are provide high defect detection rates, reduce labor costs, improveproduct quality and increase manufacturing efficiency. The automated inspection systems aregenerally consists of four phases: image acquisition, image pre-processing,feature extraction and decision making. All the phases in the system need towork in the best way to achieve the accurate and effective inspection software. 1.
2.4 Requirements for automated fabric inspection systems The maindetriments of existing inspection systems are the high cost for hardware andsoftware development, vast computational efforts are required, and the defectdetection ranges are limited. So, the researchers aim to design systems withminimal cost. The developments in fabric inspection system software facilitatefor better extraction of defects and patterns. Hence, the main requirements ofautomated inspection systems are:1) System should be designed with minimum cost.2) Automated inspection systems must provide the necessities for real-timeinspection.
3) The system must work for different types of fabric patterns.4) The defect detection rate should be higher than the human basedinspection systems.5) System should resist for poor conditions in the textile industries bothas hardware and software.
6) The system should have easy and understandable control mechanisms thatthe workers are capable of operating. 1.2.
5 Advantages of automated fabric inspection systems Automated inspection systems are designedto fulfill the requirements of the modern competitive textile industries. Inrecent years, various types of automated systems are proposed using digitalimage processing techniques as they possess more advantages than human basedinspection systems. The advantages are summarized as follows: 1) Automatedsystems increase the speed of the defect detection and also the reliability ofinspection.2) Asfabrics have very sensitive structures, the noncontact capability in automatedfabric inspection prevents the fabric pattern from disturbances that may ariseduring contact period.3) Automatedinspection systems are more stable as the effect of external conditions isless.
4) Automatedsystems provide high detection rates.