Image Processing With Artificial Intelligence

People must remember more passwords and carry more cards these days, as security becomes more important every day. However, such implementations are becoming less safe and feasible, resulting in a surge in interest in biometric systems techniques. Biometrics systems are those that record the physical characteristics of persons in an electronic environment and use that information to recognize them when needed. It works on the idea of identifying a person's physical traits that he cannot change and are distinguishable from others. The designed system aimed to successfully achieve face recognition through realtime detection, comparing the result with pre-recorded face images and tracking the human's face using a webcambased system. The system accuracy has been tested in different cases; based on the detection distance, person's appearance variation, multi-angle posing and multi-people. The system successfully implements face recognition, detection, and tracking with a satisfactory accuracy of 93%.

Ear detection is a new class of relatively stable biometrics which is not affected by facial expressions, cosmetics, eye glasses and aging effects. Ear detection is the first step of an ear recognition system, to use ear biometrics for human identification. In this paper, we have presented two approaches to detect ear from 2D side face images. One is edge detection based method and the other is template matching method. For both the methods, the correctness of the detected ear is verified using support vector machine tool. For template matching method it is also verified by Euclidian distance. The purpose of the paper is also to compare the results of both the presented methods. The experimental results prove the effectiveness of these methods.

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.

Image Segmentation plays vital role in Computer Vision and Digital Image Processing. It is the process of separating the digital image into distinct region(s) possessing homogeneous properties. The main objective of image segmentation is to extract various features of the image that are used for analyzing, interpretation and understanding of images. Image segmentation is applied in various applications like medical imaging, shape detection, content-based image retrieval, robot vision, etc. Several techniques have been developed for image segmentation such as pixel-based segmentation, edge based segmentation and region based segmentation. In this paper, segmentation technique is defined using the edge detection and morphological operations. Edge detection is done using Fuzzy Canny method for better output. After detecting the edges of image, segmentation is done using morphological operation. This gives better results.

Edge detection is one of the most commonly used operations in image analysis. Several edge detectors have been proposed in literature for enhancing and detecting of edges. In this paper a new and optimal approach of edge detection based on Cellular Automata (CA) has been proposed. The idea is simple but effective technique for edge detection that greatly improves the performances of complicated images. The comparative analysis of various image edge detection methods is presented and shown that cellular automata based algorithm performs better than all these operators under almost all scenarios.

Today, the world is going to be digitalized in all the ways. Every business units, government and private sectors, research units are using the digital image as transferring mode for every critical data. These images over the internet which will not be secure. Therefore, there is a need of image security. Currently, there exists various image security techniqueslike encryption, watermarking, steganography, etc.In this paper, we compared image encryption algorithms.Implementations of these algorithms have been realized for experimental purpose. The results of analysis are given in this paper.

Quality and execution time are two important factors for evaluation of edge detection algorithms. In these algorithms, there is a trade-off between quality and execution time. Some algorithms only concentrate on quality and some of them are fast and low quality. Efficient methods try to achieve high quality in a low time. This research concentrates on improvement of gradient based edge detection that is fast method and appropriate for real-time processing. The proposed algorithm reduces execution time by removing many pixels from computations. It calculates gradient and angle class of remaining pixels in a very efficient way so that it reinforces quality and locality of edges. The results of this algorithm indicated improvement of performance in comparison to Canny and LOG algorithms.

Today, the world is going to be digitalized in all the ways. Every business units, government and private sectors, research units are using the digital image as transferring mode for every critical data. These images over the internet which will not be secure. Therefore, there is a need of image security. Currently, there exists various image security techniqueslike encryption, watermarking, steganography, etc.In this paper, we compared image encryption algorithms.Implementations of these algorithms have been realized for experimental purpose. The results of analysis are given in this paper.

Quality and execution time are two important factors for evaluation of edge detection algorithms. In these algorithms, there is a trade-off between quality and execution time. Some algorithms only concentrate on quality and some of them are fast and low quality. Efficient methods try to achieve high quality in a low time. This research concentrates on improvement of gradient based edge detection that is fast method and appropriate for real-time processing. The proposed algorithm reduces execution time by removing many pixels from computations. It calculates gradient and angle class of remaining pixels in a very efficient way so that it reinforces quality and locality of edges. The results of this algorithm indicated improvement of performance in comparison to Canny and LOG algorithms.

Steel plates are present in submerged vessels or civil structures and usually they are in aggressive environments which can initially cause an oxidation process leading to corrosion, which can compromise their physical integrity. Periodically they undergo surveys and depending on the depth or complexity of the structure, these operations are complex and costly. The use of cameras embedded in robotic vehicles allows the use of digital image processing techniques. Oxidation has a peculiar tint which allows the use of edge detection filters. Canny filter has two input parameters, standard deviation and the thresholds that affect the identification of the edges. The purpose in this study is to choose entropy and the threshold of the image and use them as input parameters in the Canny filter to identify the oxidation points. Two images are used, one of them of a piece in which the normalized histogram focuses on dark tones and the other that has a normalized histogram concentrated in light shades.

Fractal Image Compression is a lossy compression method for images based on fractals rather than pixels, which are best suited for texture and natural images. It works on self similarity property, relying on the fact that parts of an image often resemble other parts of the same image. It takes long encoding time and affects the image quality. This paper introduces an improved quantized quad tree technique for fractal image compression. Quantized quad tree method divides the original gray level image into various blocks depending on features presented in image. Simulation results show that with the quantized quad trees compression improves the compression ratio and PSNR without losing the quality of the image compression with range block, domain block and iterations technique.

Image edge detection algorithm is applied on images to remove Gaussian noise that is present in the image during capturing or transmission using a method which combines Prewitt operator and threshold function technique to do edge detection on the image. This method is better than a method which combines Prewitt operator and mean filtering. In this paper, firstly use mean filtering to remove initially Gaussian noise, then use Prewitt operator to do edge detection on the image, and finally applied a threshold function technique with Prewitt operator.

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 .

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.

We have developed a system for automatic facial expression recognition, which runs on Google Glass and delivers real-time social cues to the wearer. We evaluate the system as a behavioral aid for children with Autism Spectrum Disorder (ASD), who can greatly benefit from real-time non-invasive emotional cues and are more sensitive to sensory input than neurotypically developing children. In addition, we present a mobile application that enables users of the wearable aid to review their videos along with auto-curated emotional information on the video playback bar. This integrates our learning aid into the context of behavioral therapy. Expanding on our previous work describing in-lab trials, this paper presents our system and application-level design decisions in depth as well as the interface learnings gathered during the use of the system by multiple children with ASD in an at-home iterative trial.

We have developed a system for automatic facial expression recognition running on Google Glass, delivering real-time social cues to children with Autism Spectrum Disorder (ASD). The system includes multiple mechanisms to engage children and their parents, who administer this technology within the home. We completed an at-home design trial with 14 families that used the learning aid over a 3-month period. We found that children with ASD generally respond well to wearing the system at home and opt for the most expressive feedback choice. We further evaluated app usage, facial engagement, and model accuracy. We found that the device can act as a powerful training aid when used periodically in the home, that interactive video content from wearable therapy sessions should be augmented with sufficient context about the content to produce long-term engagement, and that the design of wearable systems for children with ASD should be heavily dependent on the functioning level of the child. We contribute general design implications for developing wearable aids used by children with ASD and other behavioral disorders as well as their parents during at-home parent-administered therapy sessions.

Fractal Image Compression is a lossy compression method for images based on fractals rather than pixels, which are best suited for texture and natural images. It works on self similarity property, relying on the fact that parts of an image often resemble other parts of the same image. It takes long encoding time and affects the image quality. This paper introduces an improved quantized quad tree technique for fractal image compression. Quantized quad tree method divides the original gray level image into various blocks depending on features presented in image. Simulation results show that with the quantized quad trees compression improves the compression ratio and PSNR without losing the quality of the image compression with range block, domain block and iterations technique.

In this paper, an enhanced fractal image compression system (FIC) is proposed; it is based on using both symmetry prediction and blocks indexing to speed up the blocks matching process. The proposed FIC uses quad tree as variable range block partitioning mechanism. two criteria's for guiding the partitioning decision are used: The first one uses sobel-based edge magnitude, whereas the second uses the contrast of block. A new set of moment descriptors are introduced, they differ from the previously used descriptors by their ability to emphasize the weights of different parts of each block. The effectiveness of all possible combinations of double moments descriptors has been investigated. Furthermore, a fast computation mechanism is introduced to compute the moments attended to improve the overall computation cost. the results of applied tests on the system for the cases "variable and fixed range" block partitioning mechanism indicated that the variable partitioning scheme can produce better results than fixed partitioning one (that is, 4 × 4 block) in term of compression ratio, faster than and PSNR does not significantly decreased.

This paper proposes a simplified fractal image compression algorithm which is implemented on a block by block basis. This algorithm achieves a compression ratio of up to 10 with a peak signal to noise ratio (PSNR) as high as 35dB. The idea of the proposed algorithm is based on the segmentation of the image, first, into blocks to setup reference blocks. The image is then decomposed again into block ranges and a search process is carried out to find the reference blocks with best match. The transmitted or stored values, after compression, are the reference block values and the indices of the reference block that achieves the best match. If there is no match, the average value of the block range is transmitted or stored instead. The advantages of the proposed algorithm are the simplicity of computation and the high PSNR achieved.

Fractal Image Compression is a lossy compression method for images based on fractals rather than pixels, which are best suited for texture and natural images. It works on self similarity property, relying on the fact that parts of an image often resemble other parts of the same image. It takes long encoding time and affects the image quality. This paper introduces an improved quantized quad tree technique for fractal image compression. Quantized quad tree method divides the original gray level image into various blocks depending on features presented in image. Simulation results show that with the quantized quad trees compression improves the compression ratio and PSNR without losing the quality of the image compression with range block, domain block and iterations technique.

We have developed a system for automatic facial expression recognition, which runs on Google Glass and delivers real-time social cues to the wearer. We evaluate the system as a behavioral aid for children with Autism Spectrum Disorder (ASD), who can greatly benefit from real-time non-invasive emotional cues and are more sensitive to sensory input than neurotypically developing children. In addition, we present a mobile application that enables users of the wearable aid to review their videos along with auto-curated emotional information on the video playback bar. This integrates our learning aid into the context of behavioral therapy. Expanding on our previous work describing in-lab trials, this paper presents our system and application-level design decisions in depth as well as the interface learnings gathered during the use of the system by multiple children with ASD in an at-home iterative trial.

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 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. 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 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 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

One of the most important characteristics of using fractal image coding in hiding images is the encoding process that involves very long time complexity. In this paper new hiding system is proposed that hide secret image (color or gray) into cover image (color) with reducing the time of encoding process 10 times less than classical systems that use block indexing as descriptor with fractal coding. Affine transform is used in matching process to find the closer cover block for each secret block and least significant bit method to embed the coefficient affine transform. The results indicated that the encoding time is actually reduced and hiding rate reach to 100% and the quality of extracted secret image is still within the acceptable level.

The compression of medical images is essential for reducing the cost of data storage and transmission time which in turn helps in better utilization of Bandwidth. The demand for images, video sequences and computer animation has increased drastically over the years whi ch have also resulted in image and video compression. Image compression is broadly classified into lossy and lossless compression.