scispace - formally typeset
Search or ask a question

Showing papers on "Histogram equalization published in 2003"


Journal ArticleDOI
TL;DR: Simulation results show that the cases which are not handled well by HE, BBHE and dualistic sub image histogram equalization (DSIHE), can be properly enhanced by MMBEBHE.
Abstract: Histogram equalization (HE) is widely used for contrast enhancement. However, it tends to change the brightness of an image and hence, not suitable for consumer electronic products, where preserving the original brightness is essential to avoid annoying artifacts. Bi-histogram equalization (BBHE) has been proposed and analyzed mathematically that it can preserve the original brightness to a certain extends. However, there are still cases that are not handled well by BBHE, as they require higher degree of preservation. This paper proposes a novel extension of BBHE referred to as minimum mean brightness error bi-histogram equalization (MMBEBHE) to provide maximum brightness preservation. BBHE separates the input image's histogram into two based on input mean before equalizing them independently. This paper proposes to perform the separation based on the threshold level, which would yield minimum absolute mean brightness error (AMBE - the absolute difference between input and output mean). An efficient recursive integer-based computation for AMBE has been formulated to facilitate real time implementation. Simulation results using sample image which represent images with very low, very high and medium mean brightness show that the cases which are not handled well by HE, BBHE and dualistic sub image histogram equalization (DSIHE), can be properly enhanced by MMBEBHE. Besides, MMBEBHE also demonstrate comparable performance with BBHE and DSIHE when come to use the sample images show in [Yeong-Taeg Kim, February 1997] and [Yu Wan et al., October 5 1999].

853 citations


Journal ArticleDOI
TL;DR: Simulation results show that the cases which are not handled well by HE, BBHE and dualistic sub image histogram equalization (DSIHE), have been properly enhanced by RMSHE.
Abstract: Histogram equalization (HE) is widely used for contrast enhancement. However, it tends to change the brightness of an image and hence, not suitable for consumer electronic products, where preserving the original brightness is essential to avoid annoying artifacts. Bi-histogram equalization (BBHE) has been proposed and analyzed mathematically that it can preserve the original brightness to a certain extend. However, there are still cases that are not handled well by BBHE, as they require higher degree of preservation. This paper proposes a generalization of BBHE referred to as recursive mean-separate histogram equalization (RMSHE) to provide not only better but also scalable brightness preservation. BBHE separates the input image's histogram into two based on its mean before equalizing them independently. While the separation is done only once in BBHE, this paper proposes to perform the separation recursively; separate each new histogram further based on their respective mean. It is analyzed mathematically that the output image's mean brightness will converge to the input image's mean brightness as the number of recursive mean separation increases. Besides, the recursive nature of RMSHE also allows scalable brightness preservation, which is very useful in consumer electronics. Simulation results show that the cases which are not handled well by HE, BBHE and dualistic sub image histogram equalization (DSIHE), have been properly enhanced by RMSHE.

833 citations


Proceedings ArticleDOI
17 Oct 2003
TL;DR: This work investigates several illumination normalization methods and proposes some novel solutions to normalize the overall image intensity at the given illumination level.
Abstract: Evaluations of the state-of-the-art of both academic face recognition algorithms and commercial systems have shown that recognition performance of most current technologies degrades due to the variations of illumination. We investigate several illumination normalization methods and propose some novel solutions. The main contribution includes: (1) A gamma intensity correction (GIC) method is proposed to normalize the overall image intensity at the given illumination level; (2) A region-based strategy combining GIC and the histogram equalization (HE) is proposed to further eliminate the side-lighting effect; (3) A quotient illumination relighting (QIR) method is presented to synthesize images under a predefined normal lighting condition from the provided face images captured under nonnormal lighting condition. These methods are evaluated and compared on the Yale illumination face database B and Harvard illumination face database. Considerable improvements are observed. Some conclusions are given at last.

385 citations


Journal ArticleDOI
TL;DR: A new algorithm for digital images unsupervised enhancement with simultaneous global and local effects, called ACE for Automatic Color Equalization, based on a computational model of the human visual system that merges the two basic "Gray World" and "White Patch" global equalization mechanisms.

323 citations


Proceedings ArticleDOI
15 Oct 2003
TL;DR: It appears that by embedding both watermarks into one image, one could achieve extremely high robustness properties with respect to a large spectrum of image processing operations.
Abstract: The low-frequency embedding of the watermark increases the robustness with respect to image distortions that have low pass characteristics like filtering, lossy compression, geometrical distortions. On the other hand, oblivious schemes with low-frequency watermarks are more sensitive to modifications of the histogram, such as contrast/brightness adjustment, gamma correction, histogram equalization, and cropping. Watermarks inserted into middle and high frequencies are typically less robust to low-pass filtering lossy compression and small geometric deformations of the image, but are extremely robust with respect to noise adding, nonlinear deformations of the gray scale. It is understandable that the advantages and disadvantages of low and middle-to-high frequency watermarks are complementary. It appears that by embedding both watermarks into one image, one could achieve extremely high robustness properties with respect to a large spectrum of image processing operations. The above reasoning leads to proposed technique of embedding multiple watermarks into the low frequency and high frequency bands of discrete wavelet transform.

175 citations


Journal ArticleDOI
TL;DR: A facial feature extraction technique which utilizes polynomial coefficients derived from 2D Discrete Cosine Transform coefficients obtained from horizontally and vertically neighbouring blocks which is over 80 times faster to compute than features based on Gabor wavelets.

164 citations


Journal ArticleDOI
TL;DR: A novel segmentation technique, the histogram-based adaptive local thresholding (HALT), which extracts the useful information from an image without being affected by the presence of other structures, is developed, which indicates that the proposed drusen detector gives reliable detection accuracy in both position and mass size.

145 citations


Proceedings ArticleDOI
14 Dec 2003
TL;DR: In this technique, both the HSV and the YUV color spaces are incorporated to achieve better segmentation results than that of one color space techniques.
Abstract: This work is a part of a smart system that can be used in autonomous vehicles or can assist drivers in locating road signs. The detection technique includes histogram equalization, light control and color segmentation. In this technique, both the HSV and the YUV color spaces are incorporated to achieve better segmentation results than that of one color space techniques.

112 citations


Journal ArticleDOI
TL;DR: Experimental results show that the proposed color descriptor could produce a high image retrieval rate and accurately detect abrupt scene-cuts in a video analysis and the storage space required for the image histogram values can be effectively reduced.
Abstract: An important problem in color-based image retrieval and video segmentation is to lack information about how color is spatially distributed. To solve this problem and enhance the performance of image and video analyses, a spatial color descriptor is proposed involving a color adjacency histogram and color vector angle histogram. The color adjacency histogram represents the spatial distribution of color pairs at color edges in an image, thereby incorporating spatial information into the proposed color descriptor. Meanwhile, the color vector angle histogram represents the global color distribution of smooth pixels in an image. Since the proposed color descriptor includes spatial adjacency information between colors, it can robustly reduce the effect of a significant change in appearance and shape in image and video analyses. Moreover, since the color adjacency histogram is simply represented by binary streams, the storage space required for the image histogram values can be effectively reduced. Experimental results show that even with significant appearance changes, the proposed color descriptor could produce a high image retrieval rate and accurately detect abrupt scene-cuts in a video analysis.

91 citations


Journal ArticleDOI
TL;DR: A fuzzy homogeneity and scale-space approach to color image segmentation and the proposed method is compared with the space domain approach, and experimental results demonstrate the effectiveness of the proposed approach.

86 citations


Proceedings ArticleDOI
24 Nov 2003
TL;DR: This paper presents a simple enhancement rate control mechanism for the histogram equalization, which can be used to perform image processing tasks such as black/white level stretch or automatic brightness control as well as variable rate contrast enhancement.
Abstract: The histogram equalization (HE) is a widely used contrast enhancement method. But what is missing from the HE is a mechanism to control the rate of enhancement. The enhanced image always follows the uniform distribution. This paper presents a simple enhancement rate control mechanism for the HE. The gradient of the mapping function is controlled by putting constraints on the probability density function with the bin underflow (BU) and bin overflow (BO). The BUBO operation can provide the rate of enhancement from non to the full HE with a single parameter. With the enhancement rate control mechanism available, the HE can be used to perform image processing tasks such as black/white level stretch or automatic brightness control as well as variable rate contrast enhancement.

Patent
24 Feb 2003
TL;DR: In this paper, an adaptive histogram equalization method is introduced which allows the equalization amount to automatically adapt to the original image contrasts, which can be measured from the originals.
Abstract: An adaptive histogram equalization method is introduced which allows the histogram equalization amount to automatically adapt to the original image contrasts, which can be measured from the originals. Contrast over-enhancement is avoided by limiting the spatial frequency response of the histogram. Besides that, methods to remedy the brightness change problem encountered by histogram equalization are described.

Patent
30 Oct 2003
TL;DR: In this article, a method for detecting segment boundaries for a series of successive frames in a video sequence, including steps of acquiring color information from each frame, determining color histogram for each frame and applying boundary detection technique utilizing color histograms.
Abstract: The invention provides a method of detecting segment boundaries for a series of successive frames in a video sequence, including steps of acquiring color information from each frame, determining color histogram for each frame, applying boundary detection technique utilizing color histograms. Said method includes segmenting frames of video sequence into uniform color segments. Additionally, a system is provided for detecting segment boundaries for a series of successive frames in a video sequence. The system includes means for acquiring color information from each frame, means for determining color histogram for each frame, and means for applying boundary detection technique utilizing the color histograms. The system includes means for segmenting frames of video sequence into uniform color segments. Boundary detection techniques include a family color histogram method, weighted average color histogram method, successive color histogram method, stochastic method, shot-based color histogram method, mosaic color histogram method, and a computable macro-segment boundary method.

Journal ArticleDOI
TL;DR: This method involves colour quantisation, clustering and the EMD histogram difference metric to provide a transformation LUT between original and target histograms.


Journal ArticleDOI
TL;DR: The potential of a frequencybased contextual classifier (FBC) for land-use classification with a panchromatic Ikonos image is evaluated and it is found that the GLR methods resulted in accuracies similar to that of the original image, while there was little difference in classification accuracy.
Abstract: In this paper, we evaluate the potential of a frequencybased contextual classifier (FBC) for land-use classification with a panchromatic Ikonos image. To capture the spatial arrangement of image gray-level values and use such information in image classification, we applied texture spectrum (TS) directly in the FBC. The effects of several data preprocessing and reduction methods on the performance of the FBC are also evaluated. The methods include four gray-level reduction (GLR) techniques and several modifications to the TS technique. The purpose of data reduction is to improve the classification efficiency of the FBC. The GLR schemes were min-max linear compression (LC), gray level binning (BN), histogram equalization (HE), and piece-wise nonlinear compression (PC). Instead of using the texture measures derived from the texture spectrum, we directly applied texture spectra of various sizes in the classification. We modified the encoding algorithm in the TS and were able to reduce the number of texture units from its original 6561 to 256, 81, and 16, leading to as much as a 410 times computation efficiency. The original image and GLR images were subsequently classified with the FBC. We compared the classification accuracies and found that the GLR methods resulted in accuracies similar to that of the original image (within 0.03 kappa value). There was little difference in classification accuracy (within 0.03 kappa value) among the three modified TS methods, which were all outperformed by the original TS method. All TS methods performed considerably better than the use of the original image and the GLR methods.

Journal ArticleDOI
TL;DR: The proposed method is based on the concept of the chromaticity diagram and extracts a set of two-dimensional moments from it to characterize the shape and distribution of chromaticities of the given image.
Abstract: A number of different approaches have been recently presented for image retrieval using color features. Most of these methods use the color histogram or some variation of it. If the extracted information is to be stored for each image, such methods may require a significant amount of space for storing the histogram, depending on a given image's size and content. In the method proposed, only a small number of features, called chromaticity moments, are required to capture the spectral content (chrominance) of an image. The proposed method is based on the concept of the chromaticity diagram and extracts a set of two-dimensional moments from it to characterize the shape and distribution of chromaticities of the given image. This representation is compact (only a few chromaticity moments per image are required) and constant (independent of image size and content), while its retrieval effectiveness is comparable to using the full chromaticity histogram.

Proceedings ArticleDOI
24 Nov 2003
TL;DR: This paper proposes an extension of grayscale histogram equalization for color images that will always generate an almost uniform color histogram thus making an optimal use of the color space.
Abstract: In this paper we propose an extension of grayscale histogram equalization for color images. For aesthetic reasons, previously proposed color histogram equalization techniques do not generate uniform color histograms. Our method will always generate an almost uniform color histogram thus making an optimal use of the color space. This is particularly interesting for pseudo-color scientific visualization. The method is based on deforming a mesh in color space to fit the existing histogram and then map it to a uniform histogram. It is a natural extension of grayscale histogram equalization and it can be applied to spatial and color space of any dimension.

Journal ArticleDOI
TL;DR: This paper presents a color texture classification method using ratio features extracted from the color histogram by combining pairs of bins and computing corresponding count ratios that characterize the given color texture in an auto-correlative sense.

Patent
01 Apr 2003
TL;DR: In this article, an offset of a peak in a two-dimensional histogram of the colors in the representative image from a white point is used to adjust the parameters of a color correction operation according to this offset.
Abstract: Automatic color correction is applied to a scene or clip, including a sequence of images, in a motion picture by selecting a representative image of the scene, analyzing the image and adjusting parameters of a color correction operation that is performed on the sequence of images included in the scene. This operation can be repeated automatically for all scenes or for selected scenes in the motion picture. The parameters may be adjusted to automatically color balance the image while maintaining substantially constant contrast. Analysis of the representative image may include identifying an offset of a peak in a two-dimensional histogram of the colors in the representative image from a white point. Parameters of a color correction operation are adjusted according to this offset. Separate histograms and offsets may be determined for shadows, midtones and highlight regions of the representative image. Analysis of the representative image may include determining a one-dimensional histogram of the luminance information in the representative image. The darkest level and the brightest level in the image are used to balance the image. In particular, the histograms for color channels in the image, such as red, green and blue, are adjusted to match the darkest level and brightest level identified by the luminance histogram.

Patent
04 Apr 2003
TL;DR: In this paper, a method and apparatus for processing an input image to remove background color from the input image is described, which includes the step of creating a histogram to calculate a dominant color of the image and determining if a threshold luminance is less than a luminance of the dominant color.
Abstract: A method and apparatus for processing an input image to remove background color from the input image is described. The method includes the step of creating a histogram to calculate a dominant color of the input image. Next, the method determines if a threshold luminance is less than a luminance of the dominant color. If so, the method sets the output space values for RGB entries in a CMS table that match the dominant color to white. Finally, the method converts the input image to an output image by referencing the updated CMS table.

Proceedings ArticleDOI
24 Nov 2003
TL;DR: Significant modifications to the retinex approach are described, a human vision based model proposed by Land, introducing nonlinear filtering and functions to obtain improvements in both local and global contrast of the images, avoiding more complex implementations.
Abstract: New electronic imaging devices and products, such as cameras or image retouch software, demand new solutions for image correction and enhancement. In this paper we describe significant modifications to the retinex approach, a human vision based model proposed by Land, introducing nonlinear filtering and functions to obtain improvements in both local and global contrast of the images, avoiding more complex implementations. Concerning color images, we propose to work in a different color space than the classical RGB-space; in this way, color corrections, which compensate for graying-out and color shifting effects in the processed images are no longer needed.

Proceedings ArticleDOI
25 May 2003
TL;DR: Experimental results show that MHM can achieve up to 32% and up to 22% more precise than RGB-based histogram and perceptually weighted histogram methods, respectively.
Abstract: A merged histogram method (MHM) is proposed for histogram-based image retrieval based on dominant colors in images. In MHM, colors from images and between images are merged to form their dominant color set, instead of color components. Images of similar color can thus be selected and ranked. Experimental results show that MHM can achieve up to 32% and up to 22% more precise than RGB-based histogram and perceptually weighted histogram methods, respectively.

Proceedings ArticleDOI
14 Oct 2003
TL;DR: Experimental results show that an 100% accuracy of bidirectional counting can be achieved in the case of multiple isolated one-person patterns and the same accuracy can be also obtained unless the people number of a multiple-person pattern is over five.
Abstract: Based on color image processing, an automatic bidirectional counting method of pedestrians through a gate is proposed. In the developed technique, one color video camera is hung from the ceiling of the gate with a directly downward view so that the passing people will be observed from just overhead. Firstly, the passing people is roughly counted with the area of people in an image. The moving direction of the pedestrian can be oriented by tracking each people pattern through analyzing its HSI histogram. With features extracted from the quantized histograms of I (intensity) or H (hue), the first counting can be refined. Experimental results show that an 100% accuracy of bidirectional counting can be achieved in the case of multiple isolated one-person patterns and the same accuracy can be also obtained unless the people number of a multiple-person pattern is over five.

Proceedings ArticleDOI
25 May 2003
TL;DR: An object tracking method which uses back-projection of color histogram with multiple models to reduce the miss tracking even if object enters similar colored region in complex video scenes is proposed.
Abstract: Automated object tracking system is needed for unmanned observing and proper recording of important places. In this paper, we propose an object tracking method which uses back-projection of color histogram with multiple models. We can make some representative models of an object from its color histogram distribution. 3D Labeling is introduced to eliminate unsuitable histogram blobs and color histogram models are composed from survived blobs. The position of interested object could be estimated with the back-projection image of each model. The proposed method can reduce the miss tracking even if object enters similar colored region in complex video scenes.

Patent
04 Nov 2003
TL;DR: In this paper, a method for dynamically adjusting video brightness performs histogram equalization on a video YUV brightness/color signal to generate an equalization brightness curve, where the brightness of each pixel is adjusted so that the brightness difference after adjustment won't be larger than three times of the comparison difference value.
Abstract: A method for dynamically adjusting video brightness performs histogram equalization on a video YUV brightness/color signal to generate an equalization brightness curve. The brightness of each pixel is compared with every pixel of the video frame before and after histogram equalization to generate a comparison difference value. The brightness of the pixel is adjusted so that the brightness difference after adjustment won't be larger than three times of the comparison difference value and won't be smaller than a fourth of the comparison difference value. The brightness of each pixel before and after histogram equalization is compared, one by one, to generate a brightness difference value. The brightness difference value is adjusted so that the brightness difference after adjustment won't be larger than a largest brightness value and won't be smaller than a smallest brightness value. The brightness of each video frame can be properly adjusted to accomplish clearer displaying effect.

Proceedings Article
01 Jan 2003
TL;DR: A new framework of feature compensation for robust speech recognition that normalizes not only cepstral coefficients, but also their time-derivative parameters is introduced, and Delta-Cepstrum Normalization (DCN) is introduced.
Abstract: In this paper we describe a new framework of feature compensation for robust speech recognition. We introduce Delta-Cepstrum Normalization (DCN) that normalizes not only cepstral coefficients, but also their time-derivatives. In previous work, the mean and the variance of cepstral coefficients are normalized to reduce the irrelevant information, but such a normalization was not applied to time-derivative parameters because the reduction of the irrelevant information was not enough. However, Histogram Equalization provides better compensation and can be applied even to delta and delta-delta cepstra. We investigate various implementation of DCN, and show that we can achieve the best performance when the normalization of the cepstra and delta cepstra can be mutually interdependent. We evaluate the performance of DCN using speech data recorded by a PDA. DCN provides significant improvements compared to HEQ. We also examine the possibility of combining Vector Taylor Series (VTS) and DCN. Even though some combinations do not improve the performance of VTS, it is shown that the best combination gives better performance than VTS alone. Finally, the advantages of DCN in terms of the computation speed are also discussed.

Journal ArticleDOI
TL;DR: A registration method for two-dimensional (2D) registration of portal and reference images based on intensity values was tested on portal images of various anatomical sites and found that gray value based registration of portals images is applicable for a wide range of treatment sites, however, pre-processing of the images is essential.
Abstract: In external beam radiotherapy, portal imaging is applied for verification of the patient setup. Current automatic methods for portal image registration, which are often based on segmentation of anatomical structures, are especially successful for images of the pelvic region. For portal images of more complicated anatomical structures, e.g., lung, these techniques are less successful. It is desirable to have a method for image registration that is applicable for a wide range of treatment sites. In this study, a registration method for two-dimensional (2D) registration of portal and reference images based on intensity values was tested on portal images of various anatomical sites. Tests were performed with and without preprocessing (unsharp mask filtering followed by histogram equalization) for 96 image pairs and six cost functions. The images were obtained from treatments of the rectum, salivary gland, brain, prostate, and lung. To get insight into the behavior of the various cost functions, cost function values were computed for each portal image for 20,000 transformations of the corresponding reference image, translating the reference image in a range of +/- 1 cm and rotating +/- 10 degrees with respect to the clinical match. The automatic match was defined as the transformation associated with the global minimum (found by an exhaustive search). Without preprocessing, the registration reliability was low (less than 27%). With preprocessing, about 90% of the matches were successful, with a difference with our gold standard (manual registration) of about 1 mm and 1 degree SD. All tested cost functions performed similarly. However, the number of local minima using mutual information was larger than for the other tested cost functions. A cost function based on the mean product of the corresponding pixel values had the least number of local minima. In conclusion, gray value based registration of portal images is applicable for a wide range of treatment sites. However, pre-processing of the images is essential.

Proceedings ArticleDOI
TL;DR: By the simulation for the images printed by DSC, ML-MSR can improve the visibility at shadow areas keeping with both the color balance and saturation, comparing with the conventional methods, such as histogram equalization and MSR proposed by Jobson.
Abstract: With popular of Digital Still Camera-DSC, higher image quality is required. One of the subjects is that image quality at shadow area caused by the narrow dynamic range of the CCD devices is improved automatically. Conventionally, gamma transformation, histogram equalization, and etc. have been utilized for this improvement, but these are not always enough improvement. Recently, examinations applying to retinex theory taking into account of human eyes characteristics proposed by Land are paid attention. This algorithm renders image at shadow area clearly and effectively using spatial information between surrounding pixels arranged into two dimensions. Typical methods are Single-scale retinex(SSR) and Multi-Scale Retinex(MSR). These methods, however, does not always work on practical use in terms of color correction of the printed images with different RGB density distribution. In order to improve the issues of MSR, we propose the Modified Linear Multi-scale retinex (ML-MSR) method. A modified method consists of (a) linear computation processing and (b) synthesis both the original images and the images obtained by the linear MSR. By the simulation for the images printed by DSC, we show that ML-MSR can improve the visibility at shadow areas keeping with both the color balance and saturation, comparing with the conventional methods, such as histogram equalization and MSR proposed by Jobson. In general, a processing time of MSR remarkably increases with the size of Gaussian averaging filter to compute the weighted average. We describe about faster processing method of the ML-MSR algorithm, which has been shorten by using the thinning out of surrounding pixels and simplicity of average processing.

Journal ArticleDOI
TL;DR: A Scale-orientation histogram is defined for analyzing the “directionality” and “periodicity”, which are two of the most important deterministic dimensions in human texture perception.