Detail Sub-images Research Articles

Stitching defect detection and classification using wavelet transform and BP neural network

In the textile and clothing industry, much research has been conducted on fabric defect automatic detection and classification. However, little research has been done to evaluate specifically the stitching defects of a garment. In this study, a stitching detection and classification technique is presented, which combines the improved thresholding method based on the wavelet transform with the back propagation (BP) neural network. The smooth subimage at a certain resolution level using the pyramid wavelet transform was obtained. The study uses the direct thresholding method, which is based on wavelet transform smooth subimages from the use of a quadrant mean filtering method, to attenuate the texture background and preserve the anomalies. The images are then segmented by thresholding processing and noise filtering. Nine characteristic variables based on the spectral measure of the binary images were collected and input into a BP neural network to classify the sample images. The classification results demonstrate that the proposed method can identify five classes of stitching defects effectively. Comparisons of the proposed new direct thresholding method with the direct thresholding method based on the wavelet transform detailed subimages and the automatic band selection for wavelet reconstruction method were made and the experimental results show that the proposed method outperforms the other two approaches.

Image denoising algorithm using doubly local Wiener filtering with block-adaptive windows in wavelet domain

In this paper, we propose the block-adaptive windows that are used to upgrade the image denoising performance of the doubly local Wiener filtering method and the corresponding algorithm is also used to reduce spatially non-stationary additive white Gaussian noise (SNS-AWGN) in images. Based on the fact that the energy clusters in the detail subimages of an image exhibit direction features varying with spatial locations and oriented subbands, a noisy detail subimage is divided into non-overlapping small-size blocks and the spatial energy correlation function of each block is calculated to determine the principal direction of energy clusters within each block and the corresponding block-adaptive window. The block-adaptive windows are used to improve the estimations of the image's energy distribution in the detail subimages. For noisy images corrupted by SNS-AWGNs, non-uniform noise variances in the pixel domain must be estimated. To do that, we propose the joint neighborhood median absolute deviation (JNMAD) estimator, which makes the denoising algorithm able to be used in the cases of SNS-AWGNs. The experimental results show that the doubly local Wiener filtering method with block-adaptive windows is superior to other wavelet-based methods using two-dimensional separable wavelet transforms in the case of stationary noise and provides satisfactory performance in the cases of SNS-AWGNs.

Fusion of multi-spectral and panchromatic images using fuzzy rule

A fusion approach is proposed to refine the resolution of a multi-spectral image using a high-resolution panchromatic image. After the two images are decomposed by wavelet transform, five texture features are extracted from the high-frequency detailed sub-images. Then a nonlinear fusion rule, i.e. fuzzy rule is used to merge wavelet coefficients from the two images according to the extracted features. Experimental results indicate that the method outperforms the traditional approaches in preserving spectral information while improving spatial information.

Wavelet based methods on patterned fabric defect detection

The wavelet transform (WT) has been developed over 20 years and successfully applied in defect detection on plain (unpatterned) fabric. This paper is on the use of the wavelet transform to develop an automated visual inspection method for defect detection on patterned fabric. A method called direct thresholding (DT) based on WT detailed subimages has been developed. The golden image subtraction method (GIS) is also introduced. GIS is an efficient and fast method, which can segment out the defective regions on patterned fabric effectively. In this paper, the method of wavelet preprocessed golden image subtraction (WGIS) has been developed for defect detection on patterned fabric or repetitive patterned texture. This paper also presents a comparison of the three methods. It can be concluded that the WGIS method provides the best detection result. The overall detection success rate is 96.7% with 30 defect-free images and 30 defective patterned images for one common kind of patterned Jacquard fabric.

Automatic band selection for wavelet reconstruction in the application of defect detection

In this paper, we present a multiresolution approach for the inspection of local defects embedded in homogeneously textured surfaces. It is based on an efficient image restoration scheme using the wavelet transforms. By properly selecting the smooth subimage or the combination of detail subimages at different resolution levels for image reconstruction, the global repetitive texture pattern can be effectively removed and only local anomalies are preserved in the restored image. A wavelet band selection procedure is developed to automatically determine the best reconstruction parameters based on the energy distribution of wavelet coefficients. Experimental results show that the decomposed subimages and the number of resolution levels determined by the automatic band selection scheme are similar to the manual selection results, and the defects in a variety of real textures including machined surfaces, natural wood, sandpaper and textile fabrics are well detected.

Retrieving Urban Objects Using a Wavelet Transform Approach

This paper studies the use of the wavelet transform in retrieving discrete objects from urban images. The multi-scale framework of the wavelet transform method is used to represent the edge information for the retrieval of industrial complexes from digital aerial photographs. The edge information for the target objects is extracted and enhanced using a set of retrieval parameters that include summary statistical indices for the detail sub-images of the wavelet transform, an edge-pixel index, weighting schemes for these indices, the level of the wavelet transform and the search window size. Results show that all of these retrieval parameters, except for the weighting schemes, are effective in retrieving the target objects, with the majority of retrieval rates reaching 60-90 percent. Although this study focuses on industrial complexes, other types of urban objects could also be retrieved using similar retrieval framework, parameters and strategies.

Spatially adaptive multiplicative noise image denoising technique

A new image denoising technique in the wavelet transform domain for multiplicative noise is presented. Unlike most existing techniques, this approach does not require prior modeling of either the image or the noise statistics. It uses the variance of the detail wavelet coefficients to decide whether to smooth or to preserve these coefficients. The approach takes advantage of wavelet transform property in generating three detail subimages each providing specific information with certain feature directivity. This allows the ability to combine information provided by different detail subimages to direct the filtering operation. The algorithm uses the hypothesis test based on the F-distribution to decide whether detail wavelet coefficients are due to image related features or they are due to noise. The effectiveness of the proposed technique is tested for orthogonal as well as biorthogonal mother wavelets in order to study the effect of the smoothing process under different wavelet types.

Rotation-invariant pattern matching using wavelet decomposition

In this paper, we propose a wavelet decomposition approach for rotation-invariant template matching. In the matching process, we first decompose an input image into different multi-resolution levels in the wavelet-transformed domain, and use only the pixels with high wavelet coefficients in the decomposed detail subimage at a lower resolution level to compute the normalized correlation between two compared patterns. To make the matching invariant to rotation, we further use the ring-projection transform, which is invariant to object orientation, to represent an object pattern in the detail subimage. The proposed method significantly reduces the computational burden of the traditional pixel-by-pixel matching. Experimental results on a variety of real images have shown the efficacy of the proposed method.

Automatic surface inspection using wavelet reconstruction

In this paper, we present a multiresolution approach for the inspection of local defects embedded in homogeneous textured surfaces. The proposed method does not rely on the extraction of local textural features in a pixel-by-pixel basis. It is based on an efficient image restoration scheme using the wavelet transforms. By properly selecting the smooth subimage or the combination of detail subimages in different decomposition levels for backward wavelet transform, the reconstructed image will remove regular, repetitive texture patterns and enhance only local anomalies. Then, a simple thresholding can be used to discriminate between defective regions and homogeneous regions in the reconstructed image. This converts the difficult defect detection in complicated textured images into a simple binary thresholding in nontextured images. Effects of different wavelet bases, the number of multiresolution levels, the decomposed subimages used for reconstruction, and the change in image rotation on defect detection are thoroughly evaluated and discussed based on the experiment on a variety of real textures including machined surfaces, natural wood, textile fabrics, sandpaper and leather.

The Inspection of Fabric Defects by Using Wavelet Transform

In this study, wavelet transform and an artificial neural network (ANN) are used to inspect four kinds of fabric defect. Multiresolution representation of an image by using wavelet transform is a new and effective approach to the analysis of image information content. The transform can be computed efficiently by a pyramidal algorithm. The result is a set of sub-band images that consist of a lower-resolution version of the original image and a sequence of detail sub-images containing higher-spectral information. Since the transform generates localized spatial and frequency information simultaneously, the location and the kind of fabric defect can be inspected. In this study, we calculate the average and standard deviation for each sub-image as feature parameters for ANN. We explore three basic considerations on the classification rate of fabric-defect inspection consisting of wavelets with various maximum vanishing moments, different numbers of resolution levels, and different scaled-fabric images. The results show that the total classification rate for a wavelet function with a maximum vanishing moment of 4 and resolution levels of 3 can reach 100%, and the different-scaled fabric image does not obviously affect the classification rate.