Sangita Roy

In image processing technology, one of the prime applications is the dehazing process. And it is still a challenge in implementing the corresponding algorithm in hardware. In this paper we propose an atmospheric light estimation method without any complex logic. This makes easy implementation on hardware level. We have also used the Independent Transmission Rate Estimation (ITRE) for calculation of transmission coefficient parallel to the calculation of the atmospheric light. This procedure enhances the time complexity and the quality of the dehazed image. Compared to other logic studies this logic shows a competitive response.

The goal of this paper is for extricating the text present in image and identify the text written using MATLAB programming language. Text Extrication from image and recognition as a whole have enormous relevance from the point of view of application, like content-based image restoration, document indexing and identification. Moreover, this triggers vast advanced systems. This work is focusing on detecting text in an image. To find a special sign in a natural image is sometimes frustrating. By using this method, it is easy to identify this type of object in an image like traffic sign, using Optical Character Recognition (OCR). Automatic task detection algorithms help to identify such situations in an image and highly efficient. It also uses Canny Edge Detection to discover the boundaries of the text present in the image from the Maximally Stable Extremal Regions (MSER).

Image compression is a very important useful technique for efficient transmission as well as storage of images. The demand for communication of multimedia data through the telecommunication network and accessing the multimedia data through internet by utilizing less bandwidth for communication is growing explosively. Basically the image data comprise of significant portion of multimedia data and they occupy maximum portion of communication bandwidth for multimedia communication. Therefore the development of efficient image compression technique is quite necessary. The 2D Haar wavelet transform along with Hard Thresholding and Run Length Encoding is one of the efficient proposed image compression technique. JPEG2000 is a standard image compression method capable of producing very high quality compressed images. Conventional Run Length Encoding(CRLE),Optimized Run Length Encoding(ORLE),Enhanced Run Length Encoding(ERLE) are different types of RLES applied on both proposed method of compression and JPEG2000. Conventional Run Length Encoding produces efficient result for proposed method whereas Enhanced Run Length Encoding produces efficient result in JPEG2000 compression. This is the novel approach that the authors have proposed for compression of image using compression ratio (CR) without losing the PSNR, quality of image using lesser bandwidth.

In modern speech communication less bandwidth and low data rate is very essential from portable systems with limited storage capacity. Researchers are concerned with the tradeoff between bandwidth and SNR (Signal-to-Noise), BER (Bit Error Rate). Speech signal can be compressed below 64 Kbps taking care of SNR above 30 dB, and BER below 10−5. Here the authors proposed Log Normal Distribution in the design of Subband Coder and Decoder of Speech Signal taking care of the SNR and BER criterion with data rate 9.3316 Kbps.

The guided filter is derived from a local linear model. It computes the filtering output by considering the content of a guidance image (self-image or reference image). The guided filter can be used as an edge-preserving smoothing operator that behaves better near edges. It can transfer the structures of the guidance image to the filtering output, enabling new filtering applications like dehazing and guided feathering. Moreover, the guided filter naturally has a fast and non-approximate linear time algorithm, regardless of the kernel size and the intensity range. Regularization parameter (ϵ) is one of the two important factors which determines the characteristics of the guided image applied to the input image i.e. it helps to perceive the degree of the edges of an image. In this paper, we will observe the effect of the regularization parameter on a guided filtered image.