Real time fabric defect detection system on an embedded DSP platform

In this paper, we propose a novel and robust fabric defect detection method based on the low-rank representation (LRR) technique. Due to the repeated texture structure we model a defects-free fabric image as a low-rank structure. In addition, because defects, if exist, change only the texture of fabric locally, we model them with a sparse structure. Based on the above idea, we represent a fabric image into the sum of a low-rank matrix which expresses fabric texture and a sparse matrix which expresses defects. Then, the LRR method is applied to obtain the corresponding decomposition. Especially, in order to make better use of low-rank structure characteristics we propose LRREB (low-rank representation based on eigenvalue decomposition and blocked matrix) method to improve LRR. LRREB is implemented by dividing a image into some corresponding blocked matrices to reduce dimensions and applying eigen-value decomposition (EVD) on blocked matrix instead of singular value decomposition (SVD) on original fabric image, which improves the accuracy and efficiency. No training samples are required in our methods. Experimental results show that the proposed fabric defect detection method is feasible, effective, and simple to be employed.

Textile fabric defect detection based on low-rank representation

Histogram of oriented gradients (HOG) is one of the most popular descriptors used for pedestrian detection, but this descriptor has its own drawback. Like most sliding window algorithms it is very slow, making it unsuitable for many real-time applications. This paper proposes a parallel implementation of the HOG algorithm. It bases on CUDA (compute unified device architecture) platform that could use parallel computing of graphic processing unit (GPU). The time consumption of HOG running on the GPU and on the CPU is compared by experiments in this paper. The results demonstrate that the HOG on GPU performs better than the HOG running on CPU, and is approximate 10 times speedup.

Fast hog feature computation based on CUDA

Fabric inspection system is important to maintain the quality of fabric. Fabric inspection is carried out manually with human visual inspection for a long time. The work of inspectors is very tedious and consumes time and cost.To reduce the wastage of cost and time, automatic fabric inspection is required. This paper proposes an approach to recognize fabric defects in textile industry for minimizing production cost and time. The Fabric inspection system first acquires high quality vibration free images of the fabric. Then the acquired images are subjected to defect segmentation algorithm. The output of the processed image is used as an input to the Artificial Neural Network (ANN) which uses back propagation algorithm to calculate the weighted factors and generates the desired classification of defects as an output. The weaving machine is one of the easiest and fastest ways of producing cloth and textile pieces. The automated defect detection and identification system enhance the product quality and result in improved quality to meet both the customer demand and to reduce the cost associated with off-quality. This process also reduces the manual work load associated with the inspection process. Off-line monitoring system has its own disadvantages when it is compared with that of on-line monitoring system. In the off-line monitoring system the produced fabric is taken to the inspection frame where the quality of the fabric is analyzed and the fabric is transfer to the successive process. But in the on-line monitoring system the inspection of the fabric is done simultaneously while the fabric is being produced. This paper describes about fault detection, positioning and classification of the faults occur in the weaving machine during weaving by using the principle of image processing, an automatic

/pdf/fabric-inspection-system-using-artificial-neural-networks-xptgj7nm87.pdf

Fabric Inspection System using Artificial Neural Networks

Inspired by the image de-noising techniques using learned dictionaries and sparse representation, we present a fabric defect detection scheme via sparse dictionary reconstruction. Fabric defects can be regarded as local anomalies against the relatively homogeneous texture background. Following from the flexibility of sparse representation, normal fabric samples can be efficiently represented using a linear combination of a few elements of a learned dictionary. When modeling new samples with a learned dictionary, tuned to the input data containing normal fabric structural features, abnormal or defective samples are likely to have larger dissimilarity than normal samples. We evaluate the proposed methods using ten different fabric types. Experimental results show that our method has many advantages in defect detection, especially in adapting variation of fabric textures.

Sparse Dictionary Reconstruction for Textile Defect Detection

This paper proposes a textile fabric defect detection method based on the technique of matrix singular value decomposition (SVD). The matrix SVD extracts orthonormal eigen-vectors from training defect-free images of a fabric. These eigen-vectors carry structure features of the given fabric and are expressed in the format of image. For this fabric, the extracted structure features remain unchanged regardless of image sampling position as long as there is no defect. Thus the extracted structure features usually are used in projection based defect detection methods. Compared to model-based projection fabric defect detection methods, the proposed method only assumes that eigen-vectors are independent to each other. The proposed method is assessed with a number of defective fabric images. The results show that the proposed method achieves a high detection rate. In addition, factors related to such an excellent defect detection performance are also discussed.

Textile fabric flaw detection using singular value decomposition

This paper presents an efficient real-time implementation of an unsupervised textile fabric defect detection algorithm called ITT using the concept of iterative tensor tracking on graphics processing unit (GPU). The algorithm adopts a new local image descriptor, Spatial Histograms of Oriented Gradients (S-HOG), which is shift-invariant, light insensitive and space scalable. For a given textile fabric image, ITT iteratively updates and then analyzes S-HOG using tensor operations, in particular tensor decomposition to detect textile defects. To speedup the calculation required, ITT is implemented on the GPU using the Compute Unified Device Architecture (CUDA) programming model. The respective computational efficiencies of implementing ITT on the GPU and on the CPU are compared by using experiments. The results demonstrate that the computation speed of the former is on average thirty times and ten times faster than that of the later for updating the S-HOG and for detecting defects respectively because of its parallel processing nature.

K.L. Mak

Papers

Textile fabric flaw detection using singular value decomposition

Iterative tensor tracking using GPU for textile fabric defect detection