Papers by Professor Hind Rustum Mohammed
In this paper, firstly, study Hilbert matrix and some applications in image processing. Hilbert m... more In this paper, firstly, study Hilbert matrix and some applications in image processing. Hilbert matrix are used both mathematics and computational sciences ,its plays an important role in dealing with detected components based on luminance for color image, filtering the noise image and encryption the color image. And it applies in Mat lab program for any color image and any format without convert image to gray. The results proved that use Order Hilbert matrix highly efficient in the lighting effect of the components and words increased its grade Order Hilbert matrix. Whenever the light is covered on the image the concealment of the cocoons was also very clear.
One of the uncontrollable problems of the bad weather conditions is the haze. The haze causes poo... more One of the uncontrollable problems of the bad weather conditions is the haze. The haze causes poor visibility image. It attenuates the contrast and the color of the image. It creates major problems in some applications in the fields of computer vision and image processing such as intelligent vehicles, surveillance system and satellite imaging. In order to obtain the clear images, haze removal is inevitable. In this paper, we introduce a new method for haze removal. Our experiments show that the atmosphere light is the key rule for solving the problem of haze images. This paper suggested new formula to compute the atmosphere light directly, then computes two values for the atmosphere light and choose the maximum one between them. We compute the transmission map depending on the accurate value for atmosphere light which makes suggested method one of the fast methods. New formula suggested to enhance the local contrast of the image. The proposed method applied on more than 3000 haze images. The proposed method measures the quality for de-hazed images by using many quality criteria (reference quality and blind quality). The proposed method has acquired good results comparing with other methods.
The paper deals with a new way to Gray Images Deleting Components and Regions Using Symmetric gro... more The paper deals with a new way to Gray Images Deleting Components and Regions Using Symmetric group Sn. The proposed system consists of two phases, Focus on finding the edges of the image .The second stage is the image encryption phase. The system was executed on a database of 50 gray images 256 x 256 with format .png and tif. The results effectively demonstrated the proposed coding system for images using Symmetric group Sn With 100% success rate.
De-hazing image is big challenge for the researchers, although there are many good algorithms but... more De-hazing image is big challenge for the researchers, although there are many good algorithms but all of them not regards as a perfect algorithm. In this paper we try to introduce new de-hazing method based on many steps. We first preprocess the image to remove some of noise using average filter before we estimate the dark channel prior which is estimated based on average. The main contribution in this paper is the estimation of Air light value. We suggest new method based on comparing the standard deviation for RGB image and the HSV color space image. For that, we suggested many rules to control the Air light value according to experiments. Enhance local contrast implement based on V channel of color space HSV to construct visually pleasing images. The quality of de-hazed images measured visually and by using many quality measuring metrics (blind quality and reference quality) which gives promised results. Although the proposed method is not perfect method but it was more efficient than other algorithms when compared with them.
Face recognition has main attention from several foundations and researchers as a result to the i... more Face recognition has main attention from several foundations and researchers as a result to the increasing significance of security and its applications. Many approaches were introduced; each one had strengths and weaknesses. Hybrid classifier approach for face recognition using high order statistics is presented in this paper a. It consists of three stages: the first stage for features extraction which used two methods Singular Value Decomposition (SVD) and Gray Level Co-occurrence Matrix features. Hybrid method proved successfulness based on modified Structure Similarity Index (SSIM) to determine the closest similarity for face recognition results were an average of 99%. The main contributions of the proposed system represent by modifying the old ssim by Involves pasties end sill if the original function is greater than or equal to the threshold, the value of new similarity would be 1, as well as suggestion a new formula to perform classification, this formula will give good classification. Keywords:-The singular value decomposition, The gray level co-occurrence matric , Modified structure similarity index , New method for face classification, and face recognition.
Face Recognition Building new system for (face recognition problem).A system based on the integra... more Face Recognition Building new system for (face recognition problem).A system based on the integration of the following methods: _ SVD, which is used to extract three matrices, one of these matrices depends on the rows and the other on the columns and the last on the rows and columns together. The three previous matrices are considered to derive image properties. The second way is to convert the image to a single matrix. It depends on the angle. Which makes it convert the database into base properties (the person's term varies and varies with the other)? And then use a new equation for the classification. As for discrimination, it depends on the similarity that has been modified and the correlation which has also been modified by a certain threshold. We use the latest method to compile The previous methods are integrated into a new system in order to work in an integrated manner and the task of distinguishing is very excellent.
The purpose of this study encryption method will apply on all parts image by the Karhunen-Loeve T... more The purpose of this study encryption method will apply on all parts image by the Karhunen-Loeve Transform (KLT). It will be discussed time complexity and storage complexity in method (database consists of 40 images). The Karhunen-Loeve Transform (KLT) is always used in image processing and it has a wide application area. It was called "the best transform" because the decorrelation, energy concentration and Mean Square Error (MSE) is very small. The first step is found The Characteristics of the (KLT) by convert image matrix to elements vector, the second step found the covariance matrix, then compute its Great ability of decorrelation. Finally, we will reconstruct the de-nosing matrix to the Image by KLT. The experimental results and security analysis confirm the effectiveness of the indicates the robustness and advantages of the algorithm.
This work discusses the compression objects ratio for Macromedia Flash File (SWF) Image by Wavele... more This work discusses the compression objects ratio for Macromedia Flash File (SWF) Image by Wavelet functions for compression and there effect for Macromedia Flash File (SWF) Images compression. We discusses classification objects in Macromedia Flash (SWF) image in to nine types objects Action, Font, Image, Sound, Text, Button, Frame, Shape and Sprite. The work is particularly targeted towards wavelet image compression best case by using Haar Wavelet Transformation with an idea to minimize the computational requirements by applying different compression thresholds for the wavelet coefficients and these results are obtained in fraction of seconds and thus to improve the quality of the reconstructed image. The promising results obtained concerning reconstructed images quality as well as preservation of significant image details, while, on the other hand achieving high compression rates and better image quality while DB4 Wavelet Transformation higher compression rates ratio without kept for image quality .
This study deals with constructing and implementing new algorithm based on hiding a large amount ... more This study deals with constructing and implementing new algorithm based on hiding a large amount of data (image, audio, text) file into color GIF image. We have been used adaptive image filtering and adaptive image segmentation with bits replacement on the appropriate pixels. These pixels are selected randomly with Path-based animation rather than sequentially using new concept. This concept based on both visual and statistical that defined by main cases with there cases for each byte in one pixel. High security layers have been proposed through multi layers to make it difficult to break through the encryption of the input data using RC6 algorithm. The proposed algorithm can embed efficiently a large amount of data that has been reached to 86% of the image size with high quality of the output. The proposed method depended on two factors: First, the image which is containing the encrypted text, this image is GIF type which is the abbreviation of (Graphics Interchange Format Image), and the Second factor is the text targeted by encryption and entering within the image.
Colour image segmentation is an important problem in computer vision and image processing. Varian... more Colour image segmentation is an important problem in computer vision and image processing. Variant application such as image processing, computer vision, pattern recognition and machine learning widely used classical clustering method which is considered traditional k-means algorithm. K-means algorithm is famous clustering algorithm; it divided data into k clusters. The initial centroids are random selected, so the algorithm could not lead to the unique result. In this paper, we proposed a new algorithm for colour image segmentation using hybrid k-means clustering method which combine between two methods, geometric and block method. Hybrid method is to compute initial centers for k-means clustering. Geometric method depends on equal areas of distribution. Block method segments the image into uniform areas. The proposed method can overcome the drawbacks of both method (geometric and block). Furthermore, we have presented a simple validity measure based on the intra-cluster and inter-cluster distance measures which allows determining the number of clusters. The proposed method looks for the first local maximum in the validity measure. The experimental results appeared quite satisfactory.
In the present paper, a Skin pemphigus diseases image detection method color based segmentation a... more In the present paper, a Skin pemphigus diseases image detection method color based segmentation and morphological operation is proposed. Three Stages proposed: First stage: the color based segmentation takes in only one color spaces HSV, instead of three color spaces, Second stage the morphological operations with their analysis While third stage contain Extraction of connected components Skin image edge detection and a template matching. For each stage a novel algorithm which combines pixel and region based color segmentation techniques is used. Algorithm for skin segmentation of color image sequences. The experimental results for Skin pemphigus diseases image detector confirm the effectiveness of the proposed algorithm.
In this paper new algorithm for speeding up fractal image compression is presented. A new adapted... more In this paper new algorithm for speeding up fractal image compression is presented. A new adapted method based on computing the highest value of the pixel of the image to reduce the computational complexity in the encoder stage and which are led to decreasing the encoding time while the reconstructed image from the work as good as we want. For increasing the effectiveness of search stage we used another type of partitioning method that led to increase the flexibility of range partition, this method is HV-partition. We applied this method on images and also present a comparison of this method against other method which used to speed the fractal compression. Key word: fractal image compression, HV-partition, number of pixels in each block, range ,domain
In the present paper, a Skin pemphigus diseases image detection method color based segmentation a... more In the present paper, a Skin pemphigus diseases image detection method color based segmentation and morphological operation is proposed. Three Stages proposed: First stage: the color based segmentation takes in only one color spaces HSV, instead of three color spaces, Second stage the morphological operations with their analysis While third stage contain Extraction of connected components Skin image edge detection and a template matching. For each stage a novel algorithm which combines pixel and region based color segmentation techniques is used. Algorithm for skin segmentation of color image sequences. The experimental results for Skin pemphigus diseases image detector confirm the effectiveness of the proposed algorithm.
Colour image segmentation is an important problem in computer vision and image processing. Varian... more Colour image segmentation is an important problem in computer vision and image processing. Variant application such as image processing, computer vision, pattern recognition and machine learning widely used classical clustering method which is considered traditional k-means algorithm. K-means algorithm is famous clustering algorithm; it divided data into k clusters. The initial centroids are random selected, so the algorithm could not lead to the unique result. In this paper, we proposed a new algorithm for colour image segmentation using hybrid k-means clustering method which combine between two methods, geometric and block method. Hybrid method is to compute initial centers for k-means clustering. Geometric method depends on equal areas of distribution. Block method segments the image into uniform areas. The proposed method can overcome the drawbacks of both method (geometric and block). Furthermore, we have presented a simple validity measure based on the intra-cluster and inter-cluster distance measures which allows determining the number of clusters. The proposed method looks for the first local maximum in the validity measure. The experimental results appeared quite satisfactory.
— In the present paper, Mean Shift Algorithm and active contour to detect objects for CT Angiogra... more — In the present paper, Mean Shift Algorithm and active contour to detect objects for CT Angiography Image Segmentation is proposed.Based on the results we believe that this method of boundary detection together with the mean-shift can achieve fast and robust tracking of the CT Angiography Image Segmentation in noisy environment. The proposed scheme has been tested successfully on a large set of images. The performance of the proposed detector compares favorably both computationally and qualitatively, in comparison wit h Mean Shift and contour detector which are also based on surround influence .The last stage is stage contain Extraction of c onnected components CT Angiography image edge detection. The proposed scheme can serve as a low cost preprocessing step for high lev el tasks such shape based recognition and image retrieval. The experimental results confirm the effectiveness of the proposed algorithm.
In the present paper, Boundaries Object Detection for Skin Cancer Image using Connected Component... more In the present paper, Boundaries Object Detection for Skin Cancer Image using Connected Components is proposed. We propose Connected Components algorithm which that capable of Segment with Extraction of connected boundaries for Skin Cancer Image Segmentation. The algorithm is proposed to create a color label image using the local features minutiae points in skin cancer as objects image. The performance of object Detection with Connected Components which are surround influence. The proposed scheme can serve as a low cost preprocessing step for high level tasks such shape based recognition and image retrieval. The experimental results confirm the effectiveness of the proposed algorithm.
In the present paper, Arabic Character Recognition Edge detection method based on contour and con... more In the present paper, Arabic Character Recognition Edge detection method based on contour and connected components is proposed. First stage contour extraction feature is introduced to tackle the Arabic characters edge detection problem, where the aim is to extract the edge information presented in the Arabic characters, since it is crucial to understand the character content. The second stage connected components appling for the same characters to find edge detection. The proposed approach exploits a number of connected components, which move on the character by character intensity values, to establish matrix, which represents the edge information at each pixel location. The third stage the euclidean distance and vector angle are combined by using a saturation-based combination for edge detection using connected component Contour(CO3) for each character. The system has been tested on MATLAB environment with satisfactory results. Given a better device the result should increase in accuracy significantly. Fonts show that the accuracy of the proposed method is 97.4% correct characters identification in average. The contour code technique seems to be very promising producing top results. The experimental results confirm the effectiveness of the proposed algorithm.
In the present paper, algorithm is proposed to create a connected boundaries components using the... more In the present paper, algorithm is proposed to create a connected boundaries components using the local features minutiae points in image as objects image called (A. H.SH.Rostom) algorithm The Group of (A. H.SH.Rostom) algorithm consists of 10 masks were geometry of the mask operator determines a characteristic direction in which it is most sensitive to edges applied to the four stages of the Mycosis Fungoides disease Skin image have been identified and the edges of the images used for each and every stages that the database consists of 40 images divided each stage of the Mycosis Fungoides disease Skin image 10 images. The experimental results confirm the effectiveness of the proposed A. H.SH.Rostom Utilization of improved masks for Image Edge Detecting of Mycosis Fungoides.
—This paper presents Evaluation K-mean and Fuzzy c-mean image segmentation based Clustering class... more —This paper presents Evaluation K-mean and Fuzzy c-mean image segmentation based Clustering classifier. It was followed by thresholding and level set segmentation stages to provide accurate region segment. The proposed stay can get the benefits of the K-means clustering. The performance and evaluation of the given image segmentation approach were evaluated by comparing K-mean and Fuzzy c-mean algorithms in case of accuracy, processing time, Clustering classifier, and Features and accurate performance results. The database consists of 40 images executed by K-mean and Fuzzy c-mean image segmentation based Clustering classifier. The experimental results confirm the effectiveness of the proposed Fuzzy c-mean image segmentation based Clustering classifier. The statistical significance Measures of mean values of Peak signal-to-noise ratio (PSNR) and Mean Square Error (MSE) and discrepancy are used for Performance Evaluation of K-mean and Fuzzy c-mean image segmentation. The algorithm's higher accuracy can be found by the increasing number of classified clusters and with Fuzzy c-mean image segmentation.
— Credit card fraud is one of the crimes especially when it is used for web-based transaction. In... more — Credit card fraud is one of the crimes especially when it is used for web-based transaction. In this paper, a technical solution using Efficient and fast Iris authentication technique is proposed for protecting identity theft in e-commerce transactions. Therefore, this research proposes a web-based architecture which uses a combination of Image Processing and secure transmission of customers' Iris templates along with credit card details for decreasing credit card frauds over Internet. Iris image detection method based on color based segmentation and morphological operation is proposed. The color based segmentation takes in only two color spaces HSI and YCrCb, instead of three color spaces, followed by the morphological operations and a template matching. For each stage a novel algorithm which combines pixel and region based color segmentation techniques is used. The experimental results confirm the effectiveness of the proposed algorithm.
Uploads
Papers by Professor Hind Rustum Mohammed