This letter presents underwater image enhancement using Laplace decomposition. Underwater image undergoes Laplace decomposition resulting in low- and high-frequency bands. Haze is removed from the low-frequency band, and then it is normalized for white balancing. The high-frequency band is amplified for edge preservation. Adding the two frequency images results in an enhanced image. It has improved, in contrast, color, object prominence, edge preservation, reduced artifacts, and naturalness as compared to other methods related to underwater image enhancement.

This letter proposes a content-based image retrieval technique using novel dense angle descriptor and dictionary learning (DL). The histogram of oriented gradients (HOG) descriptor fails to obtain rotation invariance and well-defined rotation behavior, and therefore, a dense angle-based HOG descriptor has been presented to address the image rotation invariance. The technique computes angles across multiple scales and uses bag-of-visual features at different scales for DL. Experiments conducted on building and remote sensing datasets show that the proposed technique achieves high retrieval performance.

Abstract The paper presents an iterative matting technique for extraction of underwater objects from images. The technique adopts histogram division and stretching to obtain multiple images of different contrast levels that exhibit all image details. For each contrast image, alpha matte is produced and is further refined with every iteration. In the end, fuzzy weights are assigned to the alpha mattes obtained at different contrast levels that are combined using weighted average. The resultant alpha matte thus includes more accurate pixels from multiple alpha mattes and generates much refined matte image. The proposed technique is tested, visually and quantitatively, on a manual dataset containing 50 images. The less MSE shows that the proposed technique achieves noticeably higher accuracy as compared with contemporary image matting techniques.

This study presents a technique to automatically extract the aircraft from video sequences. The technique incorporates object tracking algorithm to separate the region of interest (ROI), i.e. movement of aircraft, from video sequence, to reduce the computational complexity. Trimaps are generated automatically for given ROI using non‐rigid image registration to obtain the prior information of the object and alpha mattes are estimated for each frame using KNN matting algorithm. However, the alpha matte of static region is estimated only once, that is combined with the alpha matte of moving region to generate final alpha matte. Simulations performed on different videos show that the proposed technique not only reduces the computational complexity, but also gives competitive results by generating high quality alpha maps.

An automatic and novel approach for acute lymphoblastic leukaemia classification is proposed. The proposed scheme is based on pre‐processing and segmentation of white blood cell nuclei using expectation maximisation algorithm, feature extraction, feature selection using principal component analysis and classification using sparse representation. The accuracy of the proposed scheme significantly outperforms the existing schemes in terms of acute lymphoblastic leukaemia classification.

The study presents a content‐based retrieval technique for retinal images. Given a query image, the aim is to automatically retrieve the relevant disease images (i.e. diabetic retinopathy, coats and choroidal neovascularisation) from the database. The proposed technique applies different pre‐processing steps to enhance the region of the exudate present in the retina. Segmentation step separates the exudates and vessels region to extract shape and texture features which are optimally combined with an automatic weight assignment approach. The simulations are performed on STARE dataset and compared with other existing techniques to ensure the significance of the proposed technique.

A framework for image visibility restoration and haze removal is proposed. The proposed technique utilizes hybrid median filtering in conjunction with accelerated local Laplacian filtering for initial dehazing of images. For visual enhancement and correct restoration of colors, constrained l0-based gradient image decomposition is applied. The proposed technique not only effectively removes haze from the images but also addresses the issues of distorted colors, visual, and halo artifacts, and haze removal from sky region in images in a better way when compared to other techniques. Experiments were performed on outdoor RGB images as well as remotely sensed images. The effectiveness of our proposed technique is demonstrated by quantitative and visual analyzes.

In this letter, an improved single image dehazing technique based on quadtree decomposition and entropy-based weighted contextual regularization is proposed. The boundary constraints are computed adaptively using statistical properties of image. The proposed technique produces high-quality dehazed image with better colors and minimal blocking artifacts. Simulation results compared visually and quantitatively with state-of-the-art existing schemes show the significance of proposed technique.

A fuzzy logic and additive wavelet-based image fusion scheme is proposed. The scheme injects high-frequency information from a high-resolution panchromatic image into a low-resolution multispectral image taking both the intensity levels of each band and spatial information of panchromatic image into consideration. The proposed scheme preserves both spatial and spectral information. Quantitative analysis performed on the Ikonos and Quickbird data sets demonstrates that the proposed scheme outperforms state-of-the-art multiresolution image fusion schemes.

A two step hybrid image fusion scheme is proposed for panchromatic and multi-spectral satellite sensors. First, we estimate an intermediate high/low resolution multi-spectral image using component substitution, which is followed by additive wavelet based high frequency injection into low resolution multi-spectral bands. Spectral dissimilarities between panchromatic and multi-spectral bands are taken into account while devising partial replacement strategy for component substitution. Quantitative analysis performed on Ikonos data set demonstrates that the proposed scheme outperforms state of the art multi-resolution image fusion schemes.

A fuzzy logic based active contour model for medical image segmentation is proposed. Image local features are incorporated in active contour model. Fuzzy logic is used to assign weights to pixels. Higher weights are assigned to pixels having less entropy and local variance whereas Lower weights are assigned to pixels having high entropy and local variance. Simulation results show the improvement in segmentation results in terms of efficiency and accuracy.

Principal component analysis and minimum description length criterion based image enhancement is proposed for through wall imaging. Minimum description length criterion is used with principal component analysis to find target eigen values. Simulation results compared on the basis of mean square error, peak signal to noise ratio and visual inspection shows the significance of proposed scheme.

A fuzzy logic and additive wavelet based fusion scheme is proposed. The scheme injects high frequency information from high resolution panchromatic image into low resolution multi-spectral bands taking both intensity level of each band and spatial information of panchromatic image into consideration. Quantitative analysis performed on IKONOS data set demonstrates that proposed scheme outperforms state of the art multi-resolution image fusion schemes.

A robust image watermarking scheme based on human visual system is proposed. The perceptual quality of the watermarked image is controlled by determining adaptive scaling factor for each individual pixel value. To determine the adaptive scaling factor, human visual system (texture masking) and fuzzy inference systems were used. The quality of the proposed scheme is tested against various attacks (Histogram equalization, rotation, Gaussian noise, scaling, cropping, JPEG compression, Y-shearing, X-shearing, median filtering, affine transformation, translation, sharpening, blurring, average filtering). The perceptual quality of watermarked image is calculated using peak-signal-to-noise-ratio, whereas the extracted watermark's quality is measured using normalized correlation.

Singular value decomposition and Fuzzy C-Means based clutter reduction is proposed for ground penetrating radar imaging. The scheme is capable of discriminating target, clutter and noise subspaces. The overlapping boundaries of clutter, noise and target signals are separated using fuzzy c-means and target image is obtained by weighted sum of different spectral components. Proposed scheme works also for extracting multiple targets in heavy cluttered images. Simulation results are compared on the basis of mean square error, peak signal to noise ratio and visual inspection.

Singular value decomposition and Akaike information criterion based image enhancement is proposed for through wall imaging. Akaike information criterion is used with singular value decomposition to find spectral images belonging to targets. Simulation results compared on the basis of mean square error, peak signal to noise ratio, miss detection, false detection and visual inspection show the significance of the proposed scheme.

Wavelet transform and principal component analysis based clutter reduction scheme is proposed for through wall imaging which is capable of discriminating between target, noise and clutter signals. Principal component analysis is used to reduce clutter and wavelet transform is used to further suppress noise signals (which results in enhancement of target signals). Proposed scheme significantly work well especially for extracting multiple targets in heavy clutters. Selection of different parameters (like mother wavelet, threshold type, threshold parameter, decomposition level and filter order etc) are explored for TWI. Existing and proposed schemes are compared on the basis of mean square error, peak signal to noise ratio and visual inspection.

Singular value decomposition based through wall image enhancement is proposed which is capable of discriminating target, noise and clutter signals. The overlapping boundaries of clutter, noise and target signals are separated using fuzzy c-means clustering algorithm. Moreover, weights are assigned to different singular values based on their membership value. Proposed scheme significantly works well for extracting multiple in heavy cluttered through wall images. Simulation results are compared on the basis mean square error, peak signal to noise ratio and visual inspection.