Md. Foisal Hossain

An image fragile watermarking scheme based on chaotic system for image tamper detection

Image watermarking techniques are specially used to provide copyright protection, owners identification, image authentication and tamper detection. Fragile watermarking is usually used for image authentication and tamper detection. In this paper, fragile watermarking scheme based on chaotic system has been proposed. Two dimensional Arnolds cat map has been used to improve the security of the proposed watermarking scheme. Arnolds cat map is sensitive to the initial values. This map is specially used to obtain the scrambled image by shuffling the pixel positions of the host image. Therefore the number of iterations and the initial values which are used to obtain scrambled image can be used as secret keys. The proposed scheme provides high security, extracts watermark from the tampered image and also localizes the tampered areas. Experimental results of proposed method have been compared with other existing two chaotic system based watermarking schemes. From the experimental results it can be seen that the proposed watermarking scheme gives better results than other chaos based watermarking schemes.

Robust driver fatigue recognition using image processing

Driver fatigue causes serious damages amongst all other road accidents. Around 20% of fatal road accidents involve driver fatigue. This paper describes a modern approach which will detect driver fatigue considering most of the fatigue symptoms. Eye closure, yawning, head tilting is the major symptoms of fatigue behavior. Inattentive vehicle movement in the road under fatigue condition is also accountable for driver fatigue. The goal of this paper is to detect these symptoms for better driving condition in the road. These symptoms are monitored by two cameras. This paper proposes a robust system where facial expressions, head tilting and lane departure for fatigue will be detected collectively. Experimental results of the proposed method are compared with the previous method. The results show good accuracy and reliable performance to avoid road accidents compared to the previous method. The proposed system is very simple and avoids any complexity.

Facial features detection in color images based on skin color segmentation

In case of intelligent biometric system facial features identification is a challenging task. Facial features like eye, mouth, lip are the critical factor to express human emotion. Face and facial features detection can be implemented automatically with the help of computer, but it is a difficult work. In this paper, a new frame work has been proposed for a fast and efficient detection of face and facial features like eyes, nose, mouth and lip from the color images. Here, the face image is given as input and facial features like eyes, mouth and lip are taken as output. The proposed algorithm increased the skin color segmentation and this method is based on three stages, face detection, region localization and facial features detection. The experimental results of the proposed method have been compared with RGB and HSI color space. The experimental results show that the proposed algorithm gives better results than the existing methods. The proposed method has eliminated the problem arise due to different pose, position of image, expressions and illumination variation problem. The average accuracy of the proposed algorithm is 97.69% and extraction of the facial features becomes easy using the proposed method from the color images.

Fast and efficient iris segmentation approach based on morphology and geometry operation

This paper presents a novel and robust method for pupil and iris boundary localization based on morphological and geometrical operation. This proposed method localizes the iris and pupil center in several stages. In the proposed algorithm Otsu threshold value, median filter, image complement, contact labeling and manual tracking are used for robust detection of the pupil. Otsu threshold is used for picking threshold value in order to determine the right binary image. Then edge point is used for the pupil boundary localization. The use of geometrical operation helps in better iris localization. Its performance does not degrade in case of seriously or partially affected eye images with eyelids and eyelashes. The proposed scheme has been tested on iris image databases like: CASIA-IrisV1 and CASIA-IrisV3. Experimental result demonstrates the supremacy of the proposed scheme in comparison with some other existing methods.

Lightning Surge Analysis on Vertical Tower using Electromagnetic Transient Program (EMTP)

This paper demonstrates the electromagnetic behaviour of transient response of lightning surge strike on vertical tower for wireless communications and power systems. Vertical tower surge responses are analysed using electromagnetic transient program (EMTP). The accuracy of this method is shown to be satisfactory with the simulation result based on the numerical electromagnetic code (NEC-2) that were carried out on vertical tower as reduced scale model. Lightning over-voltages must be taken into account when designing a tower. In the world, especially in Japan various studies, therefore, have been done to measure the characteristics of the line components and to develop their equivalent circuit models for the electromagnetic transient program (EMTP) simulations. Surge impedance is calculated by both the vertical and horizontal stroke analyses and then the single vertical and multi-storey tower surge characteristics are investigated in this work.

Classification of tumors and it stages in brain MRI using support vector machine and artificial neural network

Cell is the smallest unit of tissues, whose abnormal growth causes tumor in Brain. Support Vector Machine (SVM) and Artificial Neural Network (ANN) based tumor and its stages classification in brain MRI images is presented in this research work. This work is started with the enhancement of the brain MRI images which are obtained from oncology department of University of Maryland Medical Center. The integration of Temper based K-means and modified Fuzzy C-means (TKFCM) clustering algorithm used to segment the MRI images based on gray level intensity in small portion of brain image. The values of K in Temper based K-means algorithm more than the conventional K-means again, automatically updated membership of FCM eradicates the contouring problem of detecting the tumor region. Then, from the segmented images the first order statistic and region property based features are extracted. The first kind of features is used to detect and isolate tumor from normal brain MRI images with SVM. There is second kind which is used to classify the tumors into benign and four malignant stages tumor with ANN. The accuracy of classifying normal and tumor brain this proposed method is up to 97.37% with Bit Error Rate (BER) of 0.0294 which is better than other methods.

Tumor detection in brain MRI image using template based K-means and Fuzzy C-means clustering algorithm

This paper presents a robust segmentation method which is the integration of Template based K-means and modified Fuzzy C-means (TKFCM) clustering algorithm that, reduces operators and equipment error. In this method, the template is selected based on convolution between gray level intensity in small portion of brain image, and brain tumor image. K-means algorithm is to emphasized initial segmentation through the proper selection of template. Updated membership is obtained through distances from cluster centroid to cluster data points, until it reaches to its best. This Euclidian distance depends upon the different features i.e. intensity, entropy, contrast, dissimilarity and homogeneity of coarse image, which was depended only on similarity in conventional FCM. Then, on the basis of updated membership and automatic cluster selection, a sharp segmented image is obtained with red marked tumor from modified FCM technique. The small deviation of gray level intensity of normal and abnormal tissue is detected through TKFCM. The performances of TKFCM method is analyzed through neural network provide a better regression and least error. The performance parameters show relevant results which are effective in detecting tumor in multiple intensity based brain MRI image.

Multiple regions based histogram equalization

Image contrast enhancement is required to improve the visual appearance of the image. Histogram is an important tool for image processing. By equalizing the histogram the contrast of an image can be increased. A widely known contrast enhancement technique is histogram equalization (HE). In this paper, a histogram equalization technique has been proposed for image contrast enhancement. The input image is divided into multiple regions and each region is mapped into output region to get the equalized histogram. The proposed method is capable of preserving and maintaining the brightness and the quality of the image. The proposed method also brings out the edges which were not in the input image. Using some performance measuring tools, the proposed technique has been compared with other existing methods. From the experimental results it can be seen that the proposed technique dominates over other existing methods.

Robust RGB color image watermarking scheme based on DWT-SVD and chaotic system

This paper presents a robust image watermarking scheme for embedding gray scale watermark into RGB color image. The presented method is based on discrete wavelet transform (DWT), singular value decomposition and chaotic system. DWT and SVD are separately applied on three color planes (R, G and B) of original color image and watermark image for embedding and extracting processes. DWT and SVD techniques are used in the proposed watermarking scheme to increase the information hiding capacity and perceptual similarity of watermarked image. At the same time, chaotic system is also used in the proposed watermarking method to ensure the security of the watermarking technique by scrambling the watermark image. In the embedding stage, the gray scale watermark information is separately embedded into three color planes of the original color image. On the other hand, in the extracting stage, the watermark information is separately extracted from three color components and the final watermark is obtained by calculating the average of extracted results from three color components. The results of the proposed method are compared with other existing image watermarking methods. From the results, it can be seen that the proposed method outperforms the other existing methods by giving high values of peak signal to noise ratio (PSNR) and low values of mean square error (MSE). The proposed approach also shows well robustness against different types of attacks than other existing methods by providing high values of normalized correlation (NC).

Image watermarking approach of criminal face authentication with recovery for detecting exact criminal

This paper introduces a watermarking approach of detecting exact criminal from the received criminal photograph by criminal investigation department. As in criminal photograph, criminal face is the region of interest (ROI) part, the bits of binary logo of criminal investigation department are embedded into the criminal face as authentication information or the watermark. Before embedding, the logo is scrambled by using the generations of game of life cellular automaton. At the same time, information of criminal face is also embedded into region of non-interest (RONI) part of criminal photograph for the recovery process. Based on discrete wavelet transform (DWT), information of criminal face is embedded into the coefficients of transformed RONI. At the receiving end, at first the logo of criminal investigation department is extracted from the criminal face of received criminal photograph. If the extracted logo or authentication information is not same as the embedded logo, the information of criminal face is only then extracted from RONI. The proposed approach is evaluated by using mean squared error (MSE), peak signal to noise ratio (PSNR) and normalized correlation (NC) performance parameters. Experimental results demonstrate that the proposed approach is secure and achieves superior face recovering accuracy. In addition, the results of the proposed method are also compared with other related existing state-of-art methods. From the results of proposed watermarking approach, it can be clearly seen that the proposed approach dominates other existing methods in MSE and PSNR values.