Image surface texture analysis and classification using deep learning
- PDF / 2,072,422 Bytes
- 21 Pages / 439.37 x 666.142 pts Page_size
- 109 Downloads / 243 Views
Image surface texture analysis and classification using deep learning Akarsh Aggarwal 1 & Manoj Kumar 2 Received: 2 September 2019 / Revised: 22 June 2020 / Accepted: 21 July 2020 # Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract
Recently, the classification of surface textures is carried out using various modelling approaches. To analyse the surface texture, most of the techniques uses large amount of training data which adds up to considerable computational cost. However, the implementation of various neural network models also requires significant amount of training images to classify surface textures. In the proposed paper, a deep learning-based model is presented using convolution neural network (CNN). Further, this model is divided into two sub models knowing model-1 and model-2. The approach is designed with customized parameters configuration to classify surface texture using a smaller number of training samples. The image feature vectors are generated using statistical operations to compute the physical appearance of the surface and a CNN model is used to classify the generated surfaces with appropriate labels into classes. The Kylberg Texture dataset is used to evaluate the proposed models using 16 texture classes. The advantage of proposed models over pre-trained networks is that the entire models is customized according to specific training requirements. Further, to demonstrate the state-of-the-art results, the proposed approach is compared with other existing techniques. Our experimental results are better than the conventional techniques and achieves an accuracy of 92.42% for model-1 and 96.36% for model-2. In addition, the proposed models maintain balance between accuracy and computational cost. Keywords Surface texture classification . Deep learning . Convolution neural network . Pattern classification
* Manoj Kumar [email protected] Akarsh Aggarwal [email protected]
1
University of Bristol, Bristol, UK
2
School of Computer Science, University of Petroleum and Energy Studies (UPES), Dehradun, India
Multimedia Tools and Applications
1 Introduction Analysis of surface texture in computer vision and image processing areas is growing very fast nowadays. In fact, the texture analysis is hugely demanded in various applications. During such analysis the image parameters are used for detecting defects and flaws by providing data for recognition and interpolation of surfaces such as wood, steel, sand, fabrics, etc. It involves capturing and representation of features from surfaces of the images and further classified into appropriate classes. Generally, the extraction of surface texture is categorised/analysed using statistical approaches [28], structural approaches [29], filter based methods [22] and model based techniques. From these, the model-based approaches are widely adopted for detection and recognition of surface texture in computer vision domains. Also, the model based approaches like autoregressive models [10], fractal models, supervised learn
Data Loading...
