Recognition and classification of images is an extremely topical interdisciplinary area that covers image processing, computer graphics, artificial intelligence, computer vision, and pattern recognition, resulting in many applications based on contemporary mobile devices. Developing reliable recognition schemes is a difficult task to accomplish. It depends on many factors, such as illumination, acquisition quality and the database images, in particular, their diversity. In this paper we study how the data diversity affects decision making in image recognition, presenting a database driven classification-error predictor. The predictor is based on a hybrid approach that combines a self-organizing map together with a probabilistic logical assertion method. By means of a clustering approach, the model provides fast and efficient assessment of the image database heterogeneity and, as expected, indicates that such heterogeneity is of paramount importance for robust recognition. The practicality of the model is demonstrated using a set of image samples collected from a standard traffic sign database publicly available by the UK Department for Transport.

The quality of microscopic images is generally degraded during the image acquisition by quantizing noise, electrical noise, light illumination etc. Noise reduction is considered as a very important preprocessing step as the quality of the images can determine the accuracy of the results. The work done focuses on the noise reduction using different filters on the different types of noises applied on the common digital images and specifically the Leukemia images. 40 images were taken for the comparison purpose; 20 digital images and 20 Leukemia images of different types of Leukemia. The qualitative as well as quantitative analysis of the performance of the filters on the different noises is done. For the quantitative analysis the parameters used for the evaluation of the images are MSE, PSNR and CoC. For the qualitative analysis visual analysis in terms of quality is also done using the resultant images and their histograms. Simulation has been done in Matlab 11b. From the test cases it has been observed that Adaptive Filter produces good results on Salt and Pepper, Speckle and Gaussian noise in case of the digital images. Whereas in case of Leukemia images results of Median Filter are best for the Gaussian, Poisson and Speckle noise corrupted images.

Giving suitable input and features are always essential to obtain better accuracy in Automatic Speech Recognition (ASR). The type of signal and feature vectors given as an input is highly essential as the pattern matching algorithms strongly depends on these two components. The primary goal of this paper is to propose a suitable Pre-processing and feature extraction techniques for speaker independent speech recognition for Tamil language. The five pass Pre-processing and three types of modified feature extraction techniques are introduced using Gammatone Filtering and Cochleagram Coefficients (GFCC) to achieve better recognition performance. The modified GFCC features using multi taper Yule walker AR power spectrum, combinational features using Formant Frequencies (FF), combined frequency warping and feature normalization techniques using Linear Predictive Coding (LPC) and Cepstral Mean Normalization (CMN) are investigated. The experimental results prove that the proposed techniques have produced high recognition accuracy when compared with the conventional GFCC feature extraction technique. 

Fibromyalgia is a chronic pain syndrome that generally appears with prevalent muscular pain, sleep disorder and fatigue. Its diagnosis is very difficult. It is diagnosed in a long time after evaluating variety of psychological test scores along with physiological and laboratory tests. Psychological tests are thought not to be as reliable as laboratory test results since they depend on oral reports of the patients, and can differ from patient to patient. Beck depression inventory is one of the psychological test scores. In this study, a new biological signal that could be used instead of Beck depression inventory is introduced. For this purpose, sympathetic skin responses were used along with physiological and laboratory test results that were collected both from diagnosed fibromyalgia patients and healthy patients. A relationship based on classification was aimed to be established between the data and Beck depression inventory by using artificial neural networks. Three different artificial neural network training algorithm were used in the study. According to the results, physiological and laboratory test results and back depression inventory were estimated with the accuracy rate of 77.70\%. When all the data were used with Levenberg-Marquardt back propagation training algorithm, this rate went up to 90.91\%. According to these results, sympathetic skin responses can be safely used instead of Beck depression inventory when they were used along with other parameters that were used in fibromyalgia diagnosis.

Motion segmentation is an important task in video surveillance and in many high-level vision applications. This paper proposes two generic methods for motion segmentation from surveillance video sequences captured from different kinds of sensors like aerial, Pan Tilt and Zoom (PTZ), thermal and night vision. Motion segmentation is achieved by employing Hotelling's T-Square test on the spatial neighborhood RGB color intensity values of each pixel in two successive temporal frames. Further, a modified version of Hotelling's T-Square test is also proposed to achieve motion segmentation. On comparison with Hotelling's T-Square test, the result obtained by the modified formula is better with respect to computational time and quality of the output. Experiments along with the qualitative and quantitative comparison with existing method have been carried out on the standard IEEE PETS (2006, 2009 and 2013) and IEEE Change Detection (2014) dataset to demonstrate the efficacy of the proposed method in the dynamic environment and the results obtained are encouraging.

In every automated video surveillance system, moving object detection is an important pre-processing step leading to the extraction of useful information regarding moving objects present in a video scene. Most of the moving object detection algorithms require large memory space for storage of background related information which makes their implementation a difficult task on embedded platforms which are typically constrained by limited resources. Therefore, in order to overcome this limitation, in this paper we present a memory optimized moving object detection scheme for automated video surveillance systems with an objective to facilitate its implementation on standalone embedded platforms. The presented scheme is a modified version of the original clustering-based moving object detection algorithm and has been coded using C/C++ in the Microsoft Visual Studio IDE. The moving object detection results of the proposed memory efficient scheme were qualitatively and quantitatively analyzed and compared with the original clustering-based moving object detection algorithm. The experimental results revealed that there is 58.33% reduction in memory requirements in case of the presented memory efficient moving object detection scheme for storing background related information without any loss in accuracy and robustness as compared to the original clustering based scheme.

Human brain is a complex system, made up of neurons and glial cells. Nothing in the universe can compare with the functioning of human brain. Due to its complex nature, the diseases affected on the brain is also very complex in nature. Brain imaging is the widely used method for the diagnosing of such deceases. Brain tumor is an abnormal mass of tissue in which cells grow and multiply uncontrollably, seemingly unchecked by the mechanisms that control normal cells. Magnetic Resonance Imaging (MRI) is a commonly used modality for detecting the brain diseases. In this work we proposed a novel method for the preprocessing of MR brain images for the improved segmentation of brain tumor based on mathematical morphology operations. The first part of this paper proposes an efficient method for the skull stripping of brain MR images based on mathematical morphology. One of the main disadvantages of MRI technology is its low contrast. The second part of this paper implements an algorithm for the contrast enhancement of MR brain images using morphological operations. The output of this algorithms are evaluated using standard measures. The experimental part shows that the proposed method produces very prominent and efficient results.

An algorithm optimized towards encrypting a specific region in an image is proposed. The encryption of specific region(s) in an image is often of more practical relevance than encrypting the entire image, thus avoiding wastage of time and processing power. The algorithm proposed is ideal for encrypting images with relatively small Region of Interest (ROI). It makes novel use of the XOR operation and the relative visual redundancy of the blue-plane components in a RGB color image. A pseudorandom number generator is used as a basic security feature. Furthermore, Cipher Block Chaining (CBC) is also suggested as an improvement to further enhance the security of the algorithm. The algorithm may be classified as lossy since some visual data is sacrificed in the decrypting process. However, in case of encryption algorithms, the high security of the encrypted image is often given priority over faithful reproduction of images.