Bio-inspired approach to invariant recognition and classification of fabric weave patterns and yarn color

Abstract

Purpose This paper aims to propose a biologically inspired processing architecture to recognize and classify fabrics with respect to the weave pattern (fabric texture) and yarn color (fabric color). Design/methodology/approach By using the fabric weave patterns image identification system, this study analyzed the fabric image based on the Hierarchical-MAX (HMAX) model of computer vision, to extract feature values related to texture of fabric. Red Green Blue (RGB) color descriptor based on opponent color channels simulating the single opponent and double opponent neuronal function of the brain is incorporated in to the texture descriptor to extract yarn color feature values. Finally, support vector machine classifier is used to train and test the algorithm. Findings This two-stage processing architecture can be used to construct a system based on computer vision to recognize fabric texture and to increase the system reliability and accuracy. Using this method, the stability and fault tolerance (invariance) was improved. Originality/value Traditionally, fabric texture recognition is performed manually by visual inspection. Recent studies have proposed automatic fabric texture identification based on computer vision. In the identification process, the fabric weave patterns are recognized by the warp and weft floats. However, due to the optical environments and the appearance differences of fabric and yarn, the stability and fault tolerance (invariance) of the computer vision method are yet to be improved. By using our method, the stability and fault tolerance (invariance) was improved.

Automatic and Invariant Recognition of Traffic Sign with Color-HMAX and Deep ELM Based Classifier

“…The automatic recognition and classification accuracy of the proposed improved model is calculated and then compared with the earlier proposed model [17,[20][21] (See Table 1& 2).…”

“…In many later studies, many researchers proposed different variants of HMAX model to include color cues for the features extraction from color images and improvement in accuracy performance was reported. In our earlier work, we proposed a novel feature processing framework for the joint processing of the shape and color features based on the fusion of the HMAX model and the color opponent channels [17] and the CHMAX features output were fed to the support vector machine classifier to classify the RGB image to its respective class category. Recently, extreme learning machine was developed by Guang-Bin-Huang et al [12] as an alternative to support vector machine SVM, and it was proved to be efficient enough to compete with SVM classifiers in terms of speed, accuracy and reliability.…”

“…For a wide range of complex patterns, deep neural networks are favorable as they are more efficient in representing complex patterns with much variance due to their good generalization capability. Therefore we developed DELM deep extreme learning machine based feature classifier to further improve the recognition and classification of our earlier work [17] because it is comparatively less computation extensive and trains comparatively faster.…”

“…The output of CHMAX is then fed to the deep ELM based traffic sign classifier module as shown in Figure 1, for the classification of the traffic sign RGB image to their respective class category. Further details regarding the feature extraction module are described in detail in [17]. In this paper, we will discuss DELM extensively in the following subsection.…”

“…We chose number of non-linear neurons (M) in a hidden layer as 12,800 and number of ELM modules in cascade (D) equals 3. (see [17,19] for detailed explanation for the selection of values for optimum results)…”