Image segmentation is a growing field and it has been successfully applied in various fields such as medical imaging, face recognition, etc. In this paper, we propose a method for image segmentation that combines a region based artificial intelligence technique named fuzzy c-means (FCM) and a boundary based mathematical modeling technique level set method (LSM). In the proposed method, the contour of the image is obtained by FCM method which serves as initial contour for LSM Method. The final segmentation is achieved using LSM which uses signed pressure force (spf) function for active control of contour.

OCR of ancient document images remains a challenging task till date. Scanning process itself introduces deformation of document images. Cleaning process of these document images will result in information loss. Segmentation contributes an invariance process in OCR. Complex scripts, like derivatives of Brahmi, encounter many problems in the segmentation process. Segmentation of meaningful units, (instead of isolated patterns), revealed interesting trends. A segmentation technique for the ancient Telugu document image into meaningful units is proposed. The topological features of the meaningful units within the script line are adopted as a basis, while segmenting the text line. Horizontal profile pattern is convolved with Gaussian kernel. The statistical properties of meaningful units are explored by extensively analyzing the geometrical patterns of the meaningful unit. The efficiency of the proposed algorithm involving segmentation process is found to be 73.5% for the case of uncleaned document images.

Character Recognition is one of the important tasks in Pattern Recognition. The complexity of the character recognition problem depends on the character set to be recognized. Neural Network is one of the most widely used and popular techniques for character recognition problem. This paper discusses the classification and recognition of printed Hindi Vowels and Consonants using Artificial Neural Networks. The vowels and consonants in Hindi characters can be divided in to sub groups based on certain significant characteristics. For each group, a separate network is designed and trained to recognize the characters which belong to that group. When a test character is given, appropriate neural network is invoked to recognize the character in that group, based on the features in that character. The accuracy of the network is analyzed by giving various test patterns to the system.

The radiographic image has low contrast and high noise. In order to improve the image for observation and accurate analysis, various digital image processing techniques can be applied. In this research we propose Two Dimensional Left Median Filter method for de-noising radiographic images of welding. We have used the measures Peak Signal-to-Noise Ratio and the Mean Absolute Error for comparison. The accuracy of results obtained through our method is better than the Median and Mean Filter methods.

In this paper we have used all existing algorithms. When a fingerprint image is captured it is made pass through all the algorithms arranged in a particular order. We found that if we process a fingerprint in this particular order, the final output is good enough for minutiae detection and feature extraction. We have done many experiments on fingerprint images and found that this particular order of processing is producing better result. But for this we assume that the quality of the captured image is good enough. We have not worked on image quality enhancement. So if the input image is good our method will produce a good output. Off course this is a limitation of our proposed method, but if image is captured using a good quality device, then our method will produce an equal quality output as in other existing techniques.

Medical image processing is used for the diagnosis of diseases by the physicians or radiologists. Noise is introduced to the medical images due to various factors in medical imaging. Noise corrupts the medical images and the quality of the images degrades. This degradation includes suppression of edges, structural details, blurring boundaries etc. To diagnose diseases edge and details preservation are very important. Medical image denoising can help the physicians to diagnose the diseases. Medical images include MRI, CT scan, x-ray images, ultrasound images etc. In this paper we implemented bilateral filtering for medical image denoising. Its formulation & implementation are easy but the performance of bilateral filter depends upon its parameter. Therefore for obtaining the optimum result parameter must be estimated. We have applied bilateral filtering on medical images which are corrupted by additive white Gaussian noise with different values of variances. It is a nonlinear and local technique that preserves the features while smoothing the images. It removes the additive white Gaussian noise effectively but its performance is poor in removing salt and pepper noise.

Telemedicine is growing and there is an increased demand for faster image processing and transmitting diagnostic medical images. A region is a popular technique for image segmentation. We introduce a new approach that overcomes the close boundary initialization problem by reformulating the external energy term. We treat the contour as a mean curve of the probability density function. A widely used approach to image segmentation is to define corresponding segmentation energies and to compute shapes that are minimizes of these energies. In many medical image segmentation applications identifying and extracting the region of interest (ROI) accurately is an important step .We present a new image segmentation process, which can segment images with different image intensity distributions efficiently. To accomplish this, we construct a function that is evaluated along the evolving curve. In this cost, the value at each point on the curve is based on the analysis of interior and exterior means in a local neighborhood around that point.

An image encryption scheme based on the 3D sawtooth map is proposed in this paper. The 3D sawtooth map is utilized to generate chaotic orbits to permute the pixel positions and to generate pseudo-random gray value sequences to change the pixel gray values. The image encryption scheme is then applied to encrypt the secret image which will be imbedded in one host image. The encrypted secret image and the host image are transformed by the wavelet transform and then are merged in the frequency domain. Experimental results show that the stego-image looks visually identical to the original host one and the secret image can be effectively extracted upon image processing attacks, which demonstrates strong robustness against a variety of attacks.