Probability Distributions.- Linear Models for Regression.- Linear Models for Classification.- Neural Networks.- Kernel Methods.- Sparse Kernel Machines.- Graphical Models.- Mixture Models and EM.- Approximate Inference.- Sampling Methods.- Continuous Latent Variables.- Sequential Data.- Combining Models.

Pattern Recognition and Machine Learning

Computer and Robot Vision Vol. 1, by R.M. Haralick and Linda G. Shapiro, Addison-Wesley, 1992, ISBN 0-201-10887-1.

Computer and Robot Vision

Informative benchmarks are crucial for optimizing the page segmentation step of an OCR system, frequently the performance limiting step for overall OCR system performance. We show that current evaluation scores are insufficient for diagnosing specific errors in page segmentation and fail to identify some classes of serious segmentation errors altogether. This paper introduces a vectorial score that is sensitive to, and identifies, the most important classes of segmentation errors (over, under, and mis-segmentation) and what page components (lines, blocks, etc.) are affected. Unlike previous schemes, our evaluation method has a canonical representation of ground-truth data and guarantees pixel-accurate evaluation results for arbitrary region shapes. We present the results of evaluating widely used segmentation algorithms (x-y cut, smearing, whitespace analysis, constrained text-line finding, docstrum, and Voronoi) on the UW-III database and demonstrate that the new evaluation scheme permits the identification of several specific flaws in individual segmentation methods.

Performance Evaluation and Benchmarking of Six-Page Segmentation Algorithms

Although research in audio recognition has traditionally focused on speech and music signals, the problem of environmental sound recognition (ESR) has received more attention in recent years. Research on ESR has significantly increased in the past decade. Recent work has focused on the appraisal of non-stationary aspects of environmental sounds, and several new features predicated on non-stationary characteristics have been proposed. These features strive to maximize their information content pertaining to signal's temporal and spectral characteristics. Furthermore, sequential learning methods have been used to capture the long-term variation of environmental sounds. In this survey, we will offer a qualitative and elucidatory survey on recent developments. It includes three parts: i) basic environmental sound processing schemes, ii) stationary ESR techniques and iii) non-stationary ESR techniques. Finally, concluding remarks and future research and development trends in the ESR field will be given.

Environmental sound recognition: A survey

We develop new metrics for texture similarity that accounts for human visual perception and the stochastic nature of textures. The metrics rely entirely on local image statistics and allow substantial point-by-point deviations between textures that according to human judgment are essentially identical. The proposed metrics extend the ideas of structural similarity and are guided by research in texture analysis-synthesis. They are implemented using a steerable filter decomposition and incorporate a concise set of subband statistics, computed globally or in sliding windows. We conduct systematic tests to investigate metric performance in the context of “known-item search,” the retrieval of textures that are “identical” to the query texture. This eliminates the need for cumbersome subjective tests, thus enabling comparisons with human performance on a large database. Our experimental results indicate that the proposed metrics outperform peak signal-to-noise ratio (PSNR), structural similarity metric (SSIM) and its variations, as well as state-of-the-art texture classification metrics, using standard statistical measures.

/pdf/structural-texture-similarity-metrics-for-image-analysis-and-1urfqc48ao.pdf

Structural Texture Similarity Metrics for Image Analysis and Retrieval

We have presented a unified model for detecting different types of video shot transitions. Based on the proposed model, we formulate frame estimation scheme using the previous and the next frames. Unlike other shot boundary detection algorithms, instead of properties of frames, frame transition parameters and frame estimation errors based on global and local features are used for boundary detection and classification. Local features include scatter matrix of edge strength and motion matrix. Finally, the frames are classified as no change (within shot frame), abrupt change, or gradual change frames using a multilayer perceptron network. The proposed method is relatively less dependent on user defined thresholds and is free from sliding window size as widely used by various schemes found in the literature. Moreover, handling both abrupt and gradual transitions along with non-transition frames under a single framework using model guided visual feature is another unique aspect of the work.

A Model-Based Shot Boundary Detection Technique Using Frame Transition Parameters

We have designed and implemented an experimental CBIR system that uses a texture co-occurrence matrix. Fuzzy index of major colours are also used as colour feature to improve performance. A new measure is suggested to find out the relevance of the retrieved images and to evaluate the CBIR system. Accordingly, the performance study of the proposed system is carried out and compared with a similar system. The study has established the effectiveness of the features.

CBIR using perception based texture and colour measures

In the domain of music information retrieval, emotion based classification is an active area of research. Emotion being a perceptual and subjective concept, the task is quite challenging. It is very difficult to design signal based descriptors to represent emotions. In this work deep leaning network is proposed and experiment is done with benchmark datasets namely, Soundtracks, Bi-Modal and MER_taffc. Experiment has also been done with hand crafted descriptor consisting of different time domain and spectral features, linear predictive coding and MFCC based features. Different classifiers like, neural network, support vector machine and random forest are tried. Although the combined feature set with neural network provides an optimal result for the datasets, but in general the performance of such approaches is limited. It is difficult to obtain a consistent feature set that works across the classifier and datasets. To get rid of the issue of feature design, deep learning based approach is followed. A convolutional neural network built around VGGNet and a novel post-processing technique are proposed. Proposed methodology provides substantial improvement of performance for the datasets. Comparison with other reported works on three different datasets also establishes the superiority of the proposed methodology. The improvement in performance has been substantiated by Z test.

Recognition of emotion in music based on deep convolutional neural network

Retinal image analysis is one of the crucial topics in medical image processing. During the last three decades, people are trying to extract the different features (like blood vessels, optic disk, macula, fovea etc.) automatically from retinal image. Fovea is one of the important feature of a fundus retinal image. This paper present a simple and fast algorithm using Mathematical Morphology to find the fovea region. Proposed algorithm is based on the structure of the blood vessels and little bit information of the optic disk. We have tested our result on a publicly available DRIVE database and got a comparable results with a state of the art in this area.

http://www.isical.ac.in/~vlrg/sites/default/files/Soumitra/Paper/fundus_eait_11.pdf

A Simple and Fast Algorithm to Detect the Fovea Region in Fundus Retinal Image

In content based image retrieval (CBIR) system, search engine retrieves the images similar to the query image according to a similarity measure. It should be fast enough and must have a high precision of retrieval. Indexing scheme is used to achieve a fast response and relevance feedback helps in improving the retrieval precision. In this paper, a human perception based similarity measure is presented and based on it a simple yet novel indexing scheme with relevance feedback is discussed. The indexing scheme is designed based on the primary and secondary keys which are selected by analysing the entropy of features. A relevance feedback method is proposed based on Mann-Whitney test. The test is used to identify the discriminating features from the relevant and irrelevant images in a retrieved set. Then emphasis of the discriminating features are updated to improve the retrieval performance. The relevance feedback scheme is implemented for two different similarity measure (Euclidean distance based and human perception based). The experiment justifies the effectiveness of the proposed methodologies. Finally, the indexing scheme and relevance feedback mechanism are combined to build up the search engine.

Sanjoy Kumar Saha

Papers

A Model-Based Shot Boundary Detection Technique Using Frame Transition Parameters

CBIR using perception based texture and colour measures

Recognition of emotion in music based on deep convolutional neural network

A Simple and Fast Algorithm to Detect the Fovea Region in Fundus Retinal Image

Image retrieval based on indexing and relevance feedback