From the Publisher: 
The accessible presentation of this book gives both a general view of the entire computer vision enterprise and also offers sufficient detail to be able to build useful applications. Users learn techniques that have proven to be useful by first-hand experience and a wide range of mathematical methods. A CD-ROM with every copy of the text contains source code for programming practice, color images, and illustrative movies. Comprehensive and up-to-date, this book includes essential topics that either reflect practical significance or are of theoretical importance. Topics are discussed in substantial and increasing depth. Application surveys describe numerous important application areas such as image based rendering and digital libraries. Many important algorithms broken down and illustrated in pseudo code. Appropriate for use by engineers as a comprehensive reference to the computer vision enterprise.

Computer vision : a modern approach = 计算机视觉 : 一种现代的方法

International Journal of Computer and Applications

MUCKE aims to mine a large volume of images, to structure them conceptually and to use this conceptual structuring in order to improve large-scale image retrieval. The last decade witnessed important progress concerning low-level image representations. However, there are a number problems which need to be solved in order to unleash the full potential of image mining in applications. The central problem with low-level representations is the mismatch between them and the human interpretation of image content. This problem can be instantiated, for instance, by the incapability of existing descriptors to capture spatial relationships between the concepts represented or by their incapability to convey an explanation of why two images are similar in a content-based image retrieval framework. We start by assessing existing local descriptors for image classification and by proposing to use co-occurrence matrices to better capture spatial relationships in images. The main focus in MUCKE is on cleaning large scale Web image corpora and on proposing image representations which are closer to the human interpretation of images. Consequently, we introduce methods which tackle these two problems and compare results to state of the art methods. Note: some aspects of this deliverable are withheld at this time as they are pending review. Please contact the authors for a preview.

http://ieeexplore.ieee.org/iel2/4524/12820/00592460.pdf

Image Processing

IEEE Transactions on Pattern Analysis and Machine Intelligence

Deep Convolutional Neural Network (CNN) is a special type of Neural Networks, which has shown exemplary performance on several competitions related to Computer Vision and Image Processing. Some of the exciting application areas of CNN include Image Classification and Segmentation, Object Detection, Video Processing, Natural Language Processing, and Speech Recognition. The powerful learning ability of deep CNN is primarily due to the use of multiple feature extraction stages that can automatically learn representations from the data. The availability of a large amount of data and improvement in the hardware technology has accelerated the research in CNNs, and recently interesting deep CNN architectures have been reported. Several inspiring ideas to bring advancements in CNNs have been explored, such as the use of different activation and loss functions, parameter optimization, regularization, and architectural innovations. However, the significant improvement in the representational capacity of the deep CNN is achieved through architectural innovations. Notably, the ideas of exploiting spatial and channel information, depth and width of architecture, and multi-path information processing have gained substantial attention. Similarly, the idea of using a block of layers as a structural unit is also gaining popularity. This survey thus focuses on the intrinsic taxonomy present in the recently reported deep CNN architectures and, consequently, classifies the recent innovations in CNN architectures into seven different categories. These seven categories are based on spatial exploitation, depth, multi-path, width, feature-map exploitation, channel boosting, and attention. Additionally, the elementary understanding of CNN components, current challenges, and applications of CNN are also provided.

https://link.springer.com/content/pdf/10.1007/s10462-020-09825-6.pdf

A survey of the recent architectures of deep convolutional neural networks

Design of less-detectable RADAR waveforms using stepped frequency modulation and coding

The knowledge regarding the function of proteins is necessary as it gives a clear picture of biological processes. Nevertheless, there are many protein sequences found and added to the databases but lacks functional annotation. The laboratory experiments take a considerable amount of time for annotation of the sequences. This arises the need to use computational techniques to classify proteins based on their functions. In our work, we have collected the data from Swiss-Prot containing 40433 proteins which is grouped into 30 families. We pass it to recurrent neural network(RNN), long short term memory(LSTM) and gated recurrent unit(GRU) model and compare it by applying trigram with deep neural network and shallow neural network on the same dataset. Through this approach, we could achieve maximum of around 78% accuracy for the classification of protein families.

/pdf/deepproteomics-protein-family-classification-using-shallow-39ijha0zom.pdf

DeepProteomics: Protein family classification using Shallow and Deep Networks.

Interpretation of hyperspectral data is challenging due to the lack of spatial resolution, which causes mixing of endmember information in each pixel. Hyperspectral unmixing aims at extracting the information related to the fractional abundance of each endmember present in every pixel. The unmixing problem can be carried out by considering that the spectral signature of each endmember is a linear combination of the pure spectral signatures known in prior. In this work, sparse unmixing techniques such as, Orthogonal Matching Pursuit and Alternating Directional Multiplier Methods are applied along with dimensionality reduction of the hyperspectral image. Dimensionality reduction is obtained using the Inter-Band Block Correlation followed by singular value and QR decomposition (SVD-QR). Furthermore, we analyze the effect of dimensionality reduction on two different unmixing algorithms. Our experimentation is carried out on two real hyperspectral datasets namely ‘samson’ and ‘jasper ridge’ and the results comprises of a comparison between hyperspectral unmixing before and after dimensionality reduction using the standard metrics such as root mean square error, classwise-accuracy and visual perception. This provides a new outlook for the unmixing process as abundance estimation can be done with only the most informative bands of the image instead of using the entire data by using the dimensionality reduction technique.

Effect of Dimensionality Reduction on Sparsity Based Hyperspectral Unmixing

Robustness, Security, Imperceptibility and Data payload are the most important attributes that should be satisfied by a watermarking algorithm. A watermarking algorithm cannot satisfy all the required attributes for all applications. Best watermarking algorithms are chose, depending upon the application. Moreover, there is always a tradeoff between security versus robustness and capacity. This paper proposes a method to achieve security without the loss of robustness, imperceptibility and capacity. M-band wavelets possess the advantage of enjoying many degrees of freedom in spite of its richer parametric space. Achieving a highly secure transform domain is possible via parametric M-band wavelet transform, when the parameters used for designing are secret. This paper explains a design procedure for parametric M-band wavelets, its decomposition and reconstruction and a way to improve the security of existing robust watermarking algorithms without sacrificing any of its attributes.

Improving Security of Watermarking Algorithms via Parametric M-band Wavelet Transform

Identifying the smallest portion of the image that represents the entire image is a basic need for its efficient storage. Texture can be defined as a pattern that is repeated in a specific manner. The basic pattern that is repeated is called as Texel(Texture Element). This paper describes a method of extracting a Texel from the given textured image using K means clustering algorithm and validating it with the entire image. The number of gray levels in an image is reduced using a linear transformation function. The image is then divided in to sub windows of certain size. These sub windows are clustered together using K-means algorithm. Finally a heuristic algorithm is applied on the cluster labels to identify the Texel, which results in more than one candidate for Texel. The best among them is then chosen based on its similarity with the overall image. The similarity between the Texel and the image is calculated based on then Normalized Gray level co-occurrence matrix in the maximum gradient direction. Experiments are conducted on various texture images for various block sizes and the results are summarized.

K. P. Soman

Papers

Design of less-detectable RADAR waveforms using stepped frequency modulation and coding

DeepProteomics: Protein family classification using Shallow and Deep Networks.

Effect of Dimensionality Reduction on Sparsity Based Hyperspectral Unmixing

Improving Security of Watermarking Algorithms via Parametric M-band Wavelet Transform

Texel Identification Using K-Means Clustering Method