In this paper, we systematically review recent advances in surface inspection using computer vision and image processing techniques, particularly those based on texture analysis methods. The aim is to review the state-of-the-art techniques for the purposes of visual inspection and decision making schemes that are able to discriminate the features extracted from normal and defective regions. This field is so vast that it is impossible to cover all the aspects of visual inspection. This paper focuses on a particular but important subset which generally treats visual surface inspection as texture analysis problems. Other topics related to visual inspection such as imaging system and data acquisition are out of the scope of this survey. The surface defects are loosely separated into two types. One is local textural irregularities which is the main concern for most visual surface inspection applications. The other is global deviation of colour and/or texture, where local pattern or texture does not exhibit abnormalities. We refer this type of defects as shade or tonality problem. The second type of defects have been largely neglected until recently, particularly when colour imaging system has been widely used in visual inspection and where chromatic consistency plays an important role in quality control. The emphasis of this survey though is still on detecting local abnormalities, given the fact that majority of the reported works are dealing with the first type of defects. The techniques used to inspect textural abnormalities are discussed in four categories, statistical approaches, structural approaches, filter based methods, and model based approaches, with a comprehensive list of references to some recent works. Due to rising demand and practice of colour texture analysis in application to visual inspection, those works that are dealing with colour texture analysis are discussed separately. It is also worth noting that processing vector-valued data has its unique challenges, which conventional surface inspection methods have often ignored or do not encounter. We also compare classification approaches with novelty detection approaches at the decision making stage. Classification approaches often require supervised training and usually provide better performance than novelty detection based approaches where training is only carried out on defect-free samples. However, novelty detection is relatively easier to adapt and is particularly desirable when training samples are incomplete.

/pdf/a-review-of-recent-advances-in-surface-defect-detection-1z8rcmv73x.pdf

A Review of Recent Advances in Surface Defect Detection using Texture analysis Techniques

A Review of Quality Metrics for Fused Image

In this paper, we introduce a simple and efficient representation for natural images. We view an image (in either the spatial domain or the wavelet domain) as a collection of vectors in a high-dimensional space. We then fit a piece-wise linear model (i.e., a union of affine subspaces) to the vectors at each downsampling scale. We call this a multiscale hybrid linear model for the image. The model can be effectively estimated via a new algebraic method known as generalized principal component analysis (GPCA). The hybrid and hierarchical structure of this model allows us to effectively extract and exploit multimodal correlations among the imagery data at different scales. It conceptually and computationally remedies limitations of many existing image representation methods that are based on either a fixed linear transformation (e.g., DCT, wavelets), or an adaptive uni-modal linear transformation (e.g., PCA), or a multimodal model that uses only cluster means (e.g., VQ). We will justify both quantitatively and experimentally why and how such a simple multiscale hybrid model is able to reduce simultaneously the model complexity and computational cost. Despite a small overhead of the model, our careful and extensive experimental results show that this new model gives more compact representations for a wide variety of natural images under a wide range of signal-to-noise ratios than many existing methods, including wavelets. We also briefly address how the same (hybrid linear) modeling paradigm can be extended to be potentially useful for other applications, such as image segmentation

/pdf/multiscale-hybrid-linear-models-for-lossy-image-2o7uj0yo39.pdf

Multiscale Hybrid Linear Models for Lossy Image Representation

In this paper, a method of harmonics extraction from Higher Order Statistics (HOS) is developed for texture decomposition. We show that the diagonal slice of the fourth-order cumulants is proportional to the autocorrelation of a related noiseless sinusoidal signal with identical frequencies. We propose to use this fourth-order cumulants slice to estimate a power spectrum from which the harmonic frequencies can be easily extracted. Hence, a texture can be decomposed into deterministic components and indeterministic components as in a unified texture model through a Wold-like decomposition procedure. The simulation and experimental results demonstrated that this method is effective for texture decomposition and it performs better than traditional lower order statistics based decomposition methods.

Texture decomposition by harmonics extraction from higher order statistics

Multi-level thresholding is a helpful tool for several image segmentation applications Evaluating the optimal thresholds can be applied using a widely adopted extensive scheme called Otsu's thresholding In the current work, bi-level and multi-level threshold procedures are proposed based on their histogram using Otsu's between-class variance and a novel chaotic bat algorithm (CBA) Maximization of between-class variance function in Otsu technique is used as the objective function to obtain the optimum thresholds for the considered grayscale images The proposed procedure is applied on a standard test images set of sizes (512 × 512) and (481 × 321) Further, the proposed approach performance is compared with heuristic procedures, such as particle swarm optimization, bacterial foraging optimization, firefly algorithm and bat algorithm The evaluation assessment between the proposed and existing algorithms is conceded using evaluation metrics, namely root-mean-square error, peak signal to noise ratio, structural similarity index, objective function, and CPU time/iteration number of the optimization-based search The results established that the proposed CBA provided better outcome for maximum number cases compared to its alternatives Therefore, it can be applied in complex image processing such as automatic target recognition

Multi-level image thresholding using Otsu and chaotic bat algorithm

Texture classification by multi-model feature integration using Bayesian networks

This paper presents an adaptive texture model for texture classification. In this model, a texture is considered containing both structural and stochastic components. These two components are indeterministic and deterministic parts as in the Wold texture model that are represented by Gaussian Markov random field (GMRF) model and multichannel filtering model based on Gabor function (Gabor model), respectively. According to the different ratio of composition from each component in the texture model, an adaptive factor was proposed for the new adaptive model. Experiments demonstrated that the new adaptive model can better represent a wide variety of textures and hence can lead to better classification results.

An adaptive model for texture classification

By considering a texture being composed of two orthogonal components in a unified texture model, the deterministic component and the indeterministic component, a method of harmonics extraction from a Higher Order Statistics (HOS) based spectral decomposition is developed. The method estimates the power spectrum based on the diagonal slice of the fourth-order cumulants From this spectrum, the harmonic frequencies can be easily extracted even for noisy images. The simulation and experimental results indicate that this method is effective for texture decomposition and performs better than the traditional lower order statistics based decomposition methods.

Harmonics extraction based on higher order statistics spectrum decomposition for a unified texture model

The Wold texture model considers a textured pattern being composed of two main types of homogenous random fields, namely the deterministic and the indeterministic fields. The two fields can be represented by different type of models. It is known that, for textures, the multichannel model based on Gabor function(Gabor model) is very effective for representing the deterministic fields, and a Gaussian Markov Random Field(GMRF) model is very effective in representing the indeterministic fields. In this paper, we propose to use both models for texture classification, in which features based on each model are integrated according to the consensus theory. A weighting parameter, the deterministic energy ratio determined from the spectrum distribution function, is used as the flexible weight in the consensus theory. In this way, a wider variety of textures can be better-represented and hence lead to better classification of the textures. The classification problem is basically the problem of identifying an observed textured sample as one of several possible texture classes by a reliable but computationally attractive texture classifier. This implies that the choice of the textural features should be as compact as possible, and yet as discriminating as possible. In other words, the extraction of texture features should efficiently embody information about the textural characteristics of the image pattern. Early methods on texture classification were based on single model feature which reflected the statistical or structural properties of texture image. The statistical feature characterizes the texture by statistics of image pixel gray scale values. These methods give relatively high recognition rate when the test pattern has random looking texture, but not for large scale and more structured patterns. The structural feature assumes that the texture is generated by the placement of the primitives according to certain placement rules. These methods usually give good results in classifying structural textures[12][3]. Natural texture usually contains both structural and statistical properties. Single feature set is insufficient to describe texture image completely. Combining texture features has been suggested by some authors[10][11][9]. Conceptually, the simplest method is the stacked-vector approach in which multi-model features are concatenated together into a single feature vector and input to a classifier. This method is very straightforward and work well if the features are similar. However, the method is not applicable when the features cannot be described by a common model, e.g. the multivariate Gaussian model. In this paper, we proposed a texture classification method based on multi-model feature integration by consensus theory[1]. Two feature sets are extracted from the two components of the texture image decomposed from the image by the Wold-like decomposition[5]. Based on the consensus rule, two feature sets are combined with a flexible weight. 112 texture classes from Brodatz database[2] were used for analysis of classification performance.

/pdf/multi-model-feature-integration-for-texture-classification-43bieb0g2x.pdf

Yong Huang

Papers

Texture decomposition by harmonics extraction from higher order statistics

Texture classification by multi-model feature integration using Bayesian networks

An adaptive model for texture classification

Harmonics extraction based on higher order statistics spectrum decomposition for a unified texture model

Multi-model Feature Integration For Texture Classification