Global Histogram Equalization Research Articles

Enhancing CNN with Preprocessing Stage in Automatic Emotion Recognition

Emotion recognition from facial expression is the subfield of social signal processing which is applied in wide variety of areas, specifically for human and computer interaction. Many researches have been proposed for automatic emotion recognition, which is fundamentally using machine learning approach. However, recognizing basic emotions such as angry, happy, disgust, fear, sad, and surprise is still becoming a challenging problem in computer vision. Lately, deep learning has gained more attention to solve many real-world problems, including emotion recognition. In this research, we enhanced Convolutional Neural Network method to recognize 6 basic emotions and compared some preprocessing methods to show the influences of its in CNN performance. The compared data preprocessing methods are: resizing, face detection, cropping, adding noises, and data normalization consists of local normalization, global contrast normalization and histogram equalization. Face detection as single pre-processing phase achieved significant result with 86.08 % of accuracy, compared with another pre-processing phase and raw data. However, by combining those techniques can boost performance of CNN and achieved 97.06% of accuracy.

A Study of Transform Domain based Image Enhancement Techniques

alteration of the low complexity upgrades strategies that are in light of the singular value decomposition (SVD) for saving the mean brightness of a given picture. In spite of the fact that the SVD structured systems upgrade the reduced complexity pictures by scaling its particular worth grid, they may neglect to deliver palatable results for some low difference pictures. The weighted total of solitary lattices of the data picture and its global histogram equalization (GHE) picture is ascertained to get the particular quality framework of the leveled picture. It outflanks the routine picture evening out, for example, GHE and nearby histogram balance (LHE) and in addition the SVD systems that in light of scaling its particular quality both subjectively and quantitatively. The DWT technique can produce better quantitative measurements over the other methods. KeywordsEnhancement, DWT, DCT, SVD

An Algorithmic Approach towards Chest X-Ray Image Enhancement by Global Histogram Equalization

In medical imaging, the scope of image enhancement is highly challenging. Here digital chest x –ray image are taken in a spatial domain and enhancement of the image is done through histogram equalization method. Histogram equalization is a specific case of the more general class of histogram methods. Histogram Equalization works the best when applied to images with much higher color depth like continuos data or 16 bit gray scale images. In particular, the method can lead to better views of bone structure in X-ray images that are either over or under exposed. An algorithm is proposed to enhance the chest x-ray images using Global Histogram Equalization.

A Comparative Study between Brightness Preserving Bi-histogram and Tri-histogram Equalization for Image Contrast Enhancement

This paper compares brightness preserving image enhancement techniques using bi-histogram equalization and tri-histogram equalization methods. Traditionally for image contrast enhancement, global histogram equalization technique is used extensively. However, global histogram equalization tends to change the mean brightness of any image to the middle gray level of the dynamic range, which often results in over or under enhancement and introduce some annoying artifacts. To overcome such problems, several bi-histogram based techniques and tri-histogram based technique has been proposed. While bi-histogram based techniques divides the histogram of any image into two subhistograms and equalize them independently, tri-histogram based technique divides the histogram into three sub histograms. This paper compares some of these equalization techniques. Simulation results can be quantitative and qualitative in nature. For quantitative analysis, Absolute Mean Brightness Error (AMBE) measurement has been used. And qualitative results can be observed from the image itself. General Terms Digital Image Processing, Computer Vision

A NOVEL WRAPPING CURVELET TRANSFORMATION BASED ANGULAR TEXTURE PATTERN (WCTATP) EXTRACTION METHOD FOR WEED IDENTIFICATION

Apparently weed is a major menace in crop production as it competes with crop for nutrients, moisture, space and light which resulting in poor growth and development of the crop and finally yield. Yield loss accounts for even more than 70% when crops are frown under unweeded condition with severe weed infestation. Weed management is the most significant process in the agricultural applications to improve the crop productivity rate and reduce the herbicide application cost. Existing weed detection techniques does not yield better performance due to the complex background, illumination variation and crop and weed overlapping in the agricultural field image. Hence, there arises a need for the development of effective weed identification technique. To overcome this drawback, this paper proposes a novel Wrapping Curvelet Transformation Based Angular Texture Pattern Extraction Method (WCTATP) for weed identification. In our proposed work, Global Histogram Equalization (GHE) is used improve the quality of the image and Adaptive Median Filter (AMF) is used for filtering the impulse noise from the image. Plant image identification is performed using green pixel extraction and k-means clustering. Wrapping Curvelet transform is applied to the plant image. Feature extraction is performed to extract the angular texture pattern of the plant image. Particle Swarm Optimization (PSO) based Differential Evolution Feature Selection (DEFS) approach is applied to select the optimal features. Then, the selected features are learned and passed through an RVM based classifier to find out the weed. Edge detection and contouring is performed to identify the weed in the plant image. The Fuzzy rule-based approach is applied to detect the low, medium and high levels of the weed patchiness. From the experimental results, it is clearly observed that the accuracy of the proposed approach is higher than the existing Support Vector Machine (SVM) based approaches. The proposed approach achieves better performance in terms of accuracy.

Evaluation of various digital image processing techniques for detecting critical crescent moon and introducing CMD – A tool for critical crescent moon detection

Critical crescent moon detection is a new subject in astronomical image processing. Astronomers take photos with valuable astronomical calculations using CCD cameras and telescopes. Crescent moon might not be visible in such photos, so they must be enhanced. For this purpose, we apply common image processing methods on crescent moon images such as power operator, gamma transformation, global and local histogram equalization, POSHE, CLAHE, local enhancement, gain/offset correction, homomorphic filtering, SSR, MSR, MSRCR, wavelet and curvelet transformations. Second, we use our new methods like PB, POSP, HShistV, CurvHShistV and local brightness intensity enhancement to detect the crescent. Then we evaluate and compare the outputs of these algorithms by standard subjective evaluation tests. Finally, we conclude that these methods can give us enhanced images in which the crescent moon can be seen by human observers. We introduce our new toolbox which is named CMD, a standalone tool developed in Matlab for Critical Crescent Moon Detection using the above techniques.

Guided Image Filter and Adaptive Histogram Specification based DBLA for Enhancing Remotely Sensed Images

The fundamental reason of contrast enhancement is to bring out detail that is hidden in an image that is covered up in a picture or to expand contrast in a low contrast picture. This paper presents a literature survey on some of the image Enhancement techniques for improving images like histogram equalization, global histogram equalization, bi-histogram equalization, dualistic sub-image histogram equalization, Spatial Domain Method and Frequency Domain Method etc. The overall objective of this paper is to evaluate the short comings in earlier techniques of image enhancement.

Brightness preserving based on singular value decomposition for image contrast enhancement

This paper proposes a modification of the low contrast enhancement techniques that are based on the singular value decomposition (SVD) for preserving the mean brightness of a given image. Although the SVD-based techniques enhance the low contrast images by scaling its singular value matrix, they may fail to produce satisfactory results for some low contrast images. With the proposed method, the weighted sum of singular matrices of the input image and its global histogram equalization (GHE) image is calculated to obtain the singular value matrix of the equalized image. Simulation results show that the proposed method preserves the image brightness more precisely and enhances it with relatively negligible visual artifacts. It outperforms the conventional image equalization such as GHE and local histogram equalization (LHE), as well as the SVD techniques that based on scaling its singular value both qualitatively and quantitatively.

Exposure-based Weighted Dynamic Histogram Equalization for Image Contrast Enhancement

Global histogram equalization (GHE) [1] is a common method used for improving image contrast. However, this technique tends to introduce unnecessary visual artifacts and cannot preserve overall brightness. To overcome these problems, many ssudies have been conducted based on partitioned-histogram (i.e., sub-histogram) equalization. An input image is first divided into sub-images, individual histograms of the sub-images are equalized independently, and all of the sub-images are ultimately integrated into one complete image. For example, exposure-based sub-image histogram equalization (ESIHE) [2] uses an exposure-related threshold to divide the original image into different intensity ranges (horizontal partitioning) and also uses the mean brightness as a threshold to clip the histogram (vertical partitioning). In this paper, a novel method, called exposure-based weighted dynamic histogram equalization (EWDHE), which is an extension of ESIHE, is proposed. This study makes three major contributions to the literature. First, an Otsu-based approach and a clustering performance measure are integrated to determine the optimal number of sub-histograms and the separating points. Second, an exposure-related parameter is used to automatically adapt the contrast limitation, to avoid over-enhancement in some portions of the image. Third, a new weighted scale factor is proposed to resize the sub-histograms, which accounts for sub-histogram ranges and individual pixel numbers of these ranges. The simulation results indicated that the proposed method outperformed state-of-the-art approaches in terms of contrast enhancement, brightness preservation, and entropy preservation.

Contrast Enhancement Technique for CT Images

The images obtained by computed tomography (CT) scanning are gray scale images, in which gray level shades are almost similar around the tissue. CT images must be enhanced for easy visualization of the lesion or tumor. This paper proposes a contrast enhancement technique for medical CT images, based on modification of histogram equalization technique. Modification is done by adding a constrained variable offset to the transformation function of global histogram equalization. This constrained variable offset helps to preserve the mean brightness of the image, which prevents the global outlook of CT image. Texture of structural part of CT images was enhanced by this algorithm, due to which tumor or infected region is easily visible. The proposed technique was applied on liver CT images containing tumor and its performance was evaluated by subjective evaluation and three quantitative measures.

A Comparative Study of Histogram Equalization Techniques and Adaptive Gamma Correction for Color Image Contrast Enhancement

most noteworthy conclusion of image processing is a contrast enhancement. The most regular system for histogram equalization is utilized for retouching contrast within digital images. Histogram equalization is so advantageous and useful for image contrast enhancement method. Nonetheless, the traditional histogram equalization methods normally conclusion in surpassing complexity upgrades which variable the non-common look and noticeable curio of the handled picture. In this paper exhibits an alternate new manifestation of histogram for picture contrast enhancement. A few systems are this station is the measuring used to grant the data histogram. Global Histogram Equalization GHE utilizes the force dissemination of the whole image. Shine safeguarding Bi -Histogram Equalization BBHE utilizes the mean force is adjusted image autonomously. Double Sub -Image Histogram Equalization DSIHE utilizes the average force is adjusted image autonomously. Least Mean Brightness Error Bi -HE MMBEBHE utilizes the partition of image based on edge level, prepares the most modest Absolute Mean Brightness Error AMBE. Recursive Mean -Separate Histogram Equalization RMSHE is more different advance system for histogram adjustment. Extent Limited Bi-Histogram Equalization RLBHE jelly the initially brilliance well in order to separate the threshold that minimizes the intra - class difference. Review same that everybody these methodologies are more straightforward and valuable for image contrast

Performance Evaluation of Spatial Domain Contrast Enhancement Techniques for Underwater Images

Image enhancement methods are widely used for improving the feature and quality of the images. Researchers have designed various contrast enhancement methods in the past. But since underwater image contain low contrast and resolution due to poor visibility conditions and refractions of light rays. Thus analysis and identification of the underwater object is a challenging and difficult task. In this paper presents a comparative analysis of various contrast enhancement techniques for such underwater images. Performance of contrast limited adaptive histogram equalization (CLAHE) method is compared with contrast stretching, global histogram equalization (HE) Brightness preserving Bi-histogram equalization (BBHE) and finally Dual Sub-Image histogram equalization (DSIHE) methods. Different colour underwater images are enhanced in the RGB colour space. The performance is evaluated based on discrete entropy, signal to noise ratio (SNR) and Standard deviation difference and the mean brightness. Performance of methods is evaluated under various type of underwater image environment. General Terms Image processing, Image enhancement, Underwater Images

Multiobjectives bihistogram equalization for image contrast enhancement

The global histogram equalization (HE) has been the most frequently adopted image contrast enhancement technique. A brightness and detail‐preserving HE method with good contrast enhancement effect has been a goal of much recent research in HE. Nevertheless, producing a well‐balanced HE is deemed to be a daunting task. In this article, we propose a novel framework of HE with the aim of taking three desirable properties into account: brightness preservation, detail preservation, and contrast enhancement. We termed the proposed method as multipurpose beta optimized bi‐HE (MBOBHE). MBOBHE consists of performing the histogram optimization separately in both subhistograms after the segmentation of histogram using an optimized separating point based on the three performance criteria using a weighted‐sum aggregated objective function (AOF). Both quantitative and qualitative results indicate that MBOBHE outperforms other existing bi‐HE methods, in terms of comprehensive performance of HE that is capable of providing a holistic view. © 2014 Wiley Periodicals, Inc. Complexity 20: 22–36, 2014

Foggy Image Enhancement Based on Filter Variable Multi-Scale Retinex

In foggy weather, contrast and color of images captured by traffic visual surveillance system are degraded significantly. To improve contrast and enlarge gray range of fog-degraded image, an improved filter variable multi-scale Retinex algorithm is presented. From comparing simulation resuts by four algorithms (global histogram equalization, local histogram equalization, multiscale retinex respectively variable filter retinex), the subjective and objective evaluation is summarized respectively. Experimental results show that the method has good universality and practicality to enhance local texture details of foggy image and suppress noise effectively.

Multipurpose contrast enhancement on epiphyseal plates and ossification centers for bone age assessment

BackgroundThe high variations of background luminance, low contrast and excessively enhanced contrast of hand bone radiograph often impede the bone age assessment rating system in evaluating the degree of epiphyseal plates and ossification centers development. The Global Histogram equalization (GHE) has been the most frequently adopted image contrast enhancement technique but the performance is not satisfying. A brightness and detail preserving histogram equalization method with good contrast enhancement effect has been a goal of much recent research in histogram equalization. Nevertheless, producing a well-balanced histogram equalized radiograph in terms of its brightness preservation, detail preservation and contrast enhancement is deemed to be a daunting task.MethodIn this paper, we propose a novel framework of histogram equalization with the aim of taking several desirable properties into account, namely the Multipurpose Beta Optimized Bi-Histogram Equalization (MBOBHE). This method performs the histogram optimization separately in both sub-histograms after the segmentation of histogram using an optimized separating point determined based on the regularization function constituted by three components. The result is then assessed by the qualitative and quantitative analysis to evaluate the essential aspects of histogram equalized image using a total of 160 hand radiographs that are implemented in testing and analyses which are acquired from hand bone online database.ResultFrom the qualitative analysis, we found that basic bi-histogram equalizations are not capable of displaying the small features in image due to incorrect selection of separating point by focusing on only certain metric without considering the contrast enhancement and detail preservation. From the quantitative analysis, we found that MBOBHE correlates well with human visual perception, and this improvement shortens the evaluation time taken by inspector in assessing the bone age.ConclusionsThe proposed MBOBHE outperforms other existing methods regarding comprehensive performance of histogram equalization. All the features which are pertinent to bone age assessment are more protruding relative to other methods; this has shorten the required evaluation time in manual bone age assessment using TW method. While the accuracy remains unaffected or slightly better than using unprocessed original image. The holistic properties in terms of brightness preservation, detail preservation and contrast enhancement are simultaneous taken into consideration and thus the visual effect is contributive to manual inspection.

Major Component Suppression Histogram Equalization for Medical Image Enhancement

Histogram equalization technique is commonly and widely applied for image enhancement. However, this approach is not suitable for medical images that have the major component of background. It may cause a wash-out effect after histogram equalization for image enhancement. In this paper, a modified histogram equalization algorithm called major component suppression histogram equalization (MCSHE for short) was proposed and demonstrated. In this proposed method, the gray-level of highest probability is considered as background and masked out before applying the traditional global histogram equalization process. Afterwards, a new gray value is interpolated for histogram transformation. From the experimental results, the proposed method is demonstrated to show the excellent image quality in enhancement for visualization without introducing wash-out appearance. Thus, this modified histogram equalization for medical image enhancement without the wash-out effect is helpful and useful in medical applications.

Semi-Automatic Global Contrast Enhancement

local contrast enhancement is not sufficient for a detailed visibility of an image, global enhancement also cannot be neglected. Among the global contrast enhancement methods, the automatic enhancement method like Global Histogram Equalization is not always desirable to utilize it for some images that some portions are overexposure and some portions are underexposure and no user's choice is available. And for other global enhancement methods, when the number of user defined parameters are more, more number of different choices are available and the enhancement level can be adjusted more accurately. However, the user's convenience is less when more number of user defined parameters. For the sake of user's convenience, the semi-automatic contrast enhancement method using single user defined parameter works better for some images.

가우시안 혼합 모델 기반의 영상 히스토그램 평활화

영상에서 밝기 분포가 특정한 범위에 밀집되어 있는 경우 영상에 포함된 특징을 구분하기가 어렵다. 이러한 문제를 해결하기 위해서 전역 히스토그램 평활화와 지역 히스토그램 평활화를 적용한다. 전역 히스토그램 평활화를 적용하는 경우 밝기 분포의 밀집 정도를 고려하지 않고 전체 히스토그램 정보를 사용하기 때문에 지나치게 밝아지거나 어두워질 수 있으며 부분적인 명암값을 개선시키는 것이 어렵다. 지역 히스토그램 평활화를 적용하는 경우 영상의 전체 밝기 분포를 고려하지 않고 지역적인 영상의 밝기 정보만을 사용하기 때문에 블록 간의 명암값의 차가 커져서 블록화 현상이 발생한다. 이러한 문제를 해결하기 위해 본 논문에서는 영상의 히스토그램의 영역에 가우시안 혼합 모델을 적용하여 모델링을 한 후, EM 알고리즘을 반복적으로 적용하여 각 영역의 범위를 결정한다. 그리고 분할된 영역별로 히스토그램 평활화를 적용하여 유사한 밝기값을 갖는 영역이 과도하게 평활화 되는 것을 방지하며 명암대비를 향상시킨다. In case brightness distribution is concentrated in a region, it is difficult to classify the image features. To solve this problem, we apply global histogram equalization and local histogram equalization to images. In case of global histogram equalization, it can be too bright or dark because it doesn't consider the density of brightness distribution. Thus, it is difficult to enhance the local contrast in the images. In case of local histogram equalization, it can produce unexpected blocks in the images. In order to enhance the contrast in the images, this paper proposes a local histogram equalization based on the Gaussian Mixture Models(GMMs) in regions of histogram. Mean and variance parameters in each regions is updated EM-algorithm repeatedly and then ranges of equalization on each regions. The experimental results performed with image of various contrasts show that the proposed algorithm is better than the global histogram equalization.

An adaptive image enhancement algorithm

Image enhancement is one of the most important techniques in digital image processing field. Traditional global histogram equalization usually causes excessive contrast enhancement while local histogram equalization may cause block effect. So a new image enhancement technique based on human visual properties is proposed which improves contrast and edges in an image as well as keeps the noise level lower.

Two-dimensional histogram equalization and contrast enhancement

In this paper, we propose a two-dimensional histogram equalization (2DHE) algorithm which utilizes contextual information around each pixel to enhance the contrast of an input image. The algorithm is based on the observation that the contrast in an image can be improved by increasing the grey-level differences between each pixel and its neighbouring pixels. The image equalization is achieved by assuming that for a given image, the modulus of the grey-level differences between pixels and their neighbouring pixels are equally distributed. The well-known global histogram equalization algorithm is a special case of 2DHE when contextual information is not utilized. 2DHE is easy to implement requiring only a small number of simple arithmetic operations and is thus suitable for real-time contrast enhancement applications. Experimental results show that 2DHE produces better or comparable enhanced images than several state-of-the-art algorithms. The only parameter in 2DHE which requires tuning is the size of the spatial neighbourhood support which provides the contextual information for a given dynamic range of the enhanced image. An automated parameter selection algorithm is also presented. The algorithm can be applied to a wide range of image types.