Abstract

IMAGE RESTORATION USING KNOWLEDGE FROM THE IMAGE

Abstract  There are various real world situations where, a portion of the image is lost or damaged which needs an image restoration. A Prior knowledge of the image may not be available for restoring the image, which demands for a knowledge derivation from the image itself. Restoring the lost portions of the image based on the knowledge obtained from the image area surrounding the lost area is called as Digital Image Inpainting. The information content in the lost area could contain structural information like edges or textural information like repeating patterns. This knowledge is derived from the boundary area surrounding the lost area. Based on this, the lost area is restored by looking at similar information in the same image. Experimentation have been done on various images and observed that the algorithm restores the image in a visually plausible way. 

COMPARATIVE ANALYSIS OF STRUCTURE AND TEXTURE BASED IMAGE INPAINTING TECHNIQUES

Abstract— There are various real world situations where, a portion of the image is lost or damaged or hidden by an unwanted object which needs an image restoration. Digital Image Inpainting is a technique which addresses such an issue. Inpainting techniques are based on interpolation, diffusion or exemplar based concepts. This paper briefly describes the application of such concepts for inpainting and provides their detailed performance analysis. It is observed that the performance of these techniques vary while restoring the structure and texture present in an image. This paper gives the limitations of each technique and suggests the choice of appropriate technique for a given scenario. 

A HIERARCHICAL SEARCH SPACE REFINEMENT AND FILLING FOR EXEMPLAR BASED IMAGE INPAINTING

Abstract- There are many real world scenarios where a portion of the image is damaged or lost. Restoring such an image without prior knowledge or a reference image is a difficult task. Image inpainting is a method that focuses on reconstructing the damaged or missing portion of images based on the information available from undamaged areas of the same image. The existing methods fill the missing area from the boundary. Their performance varies while reconstructing structures and textures and many of them restrict the size of the area to be inpainted. In this paper exemplar based inpainting is adopted in a hierarchical framework. A hierarchical search space refinement and hierarchical filling are proposed in this paper which increases the accuracy and handles the extra cost due to multi resolution processing in a better way. The former tries to select an exemplar suitable at all resolution levels restricting the search space from the lower resolution level. The later fills the region at lower resolution level whose results are taken to the higher levels. This makes the non boundary pixels known in the higher resolution level which in turn helps in search space refinement while increasing accuracy.

HIERARCHICAL APPROACH FOR TOTAL VARIATION DIGITAL IMAGE INPAINTING

Abstract —The art of recovering an image from damage in an undetectable form is known as inpainting. The manual work of inpainting is most often a very time consuming process. Due to digitalization of this technique, it is automatic and faster. In this paper, after the user selects the regions to be reconstructed, the algorithm automatically reconstruct the lost regions with the help of the information surrounding them. The existing methods perform very well when the region to be reconstructed is very small, but fails in proper reconstruction as the area increases. This paper describes a Hierarchical method by which the area to be inpainted is reduced in multiple levels and Total Variation(TV) method is used to inpaint in each level. This algorithm gives better performance when compared with other existing algorithms such as nearest neighbor interpolation, Inpainting through Blurring and Sobolev Inpainting.

LAPLACIAN PYRAMID BASEDHIERARCHICAL IMAGE INPAINTING

Abstract - There are many real world scenarios where a portion of the image is damaged or lost. Restoring such an image without prior knowledge or a reference image is a difficult task. Image inpainting is a method that focuses on reconstructing the damaged or missing portion of images based on the information available from undamaged areas of the same image. The existing methods fill the missing area from the boundary. Their performance varies while reconstructing the structure and texture present in the image and majorly fails for larger inpainting area. This paper attempts to segregate the structure and texture using Laplacian Pyramid and inpaint them separately using a top down approach. The images are inpainted from the lowest spatial resolution using Exemplar based image synthesis. The results are updated before moving to the higher resolution levels. This multi resolution process ensures the coarser details being filled before the finer details. The structure propagation is better since it is handled separately. The top down approach alleviates the traditional boundary based filling and breaks the single large sized inpainting region into many smaller sized ones as we move down the pyramid. Different types of images have been experimented and the results are summarized.

Hierarchical Digital Image Inpainting Using Wavelets

Abstract-Inpainting is the technique of reconstructing unknown or damaged portions of an image in a visually plausible way. Inpainting algorithm automatically fills the damaged region in an image using the information available in undamaged region. Propagation of structure and texture information becomes a challenge as the size of damaged area increases. In this paper, a hierarchical inpainting algorithm using wavelets is proposed. The hierarchical method tries to keep the mask size smaller while wavelets help in handling the high pass structure information and low pass texture information separately. The performance of the proposed algorithm is tested using different factors. The results of our algorithm are compared with existing methods such as interpolation, diffusion and exemplar techniques.

STRUCTURAL INFORMATION OF IMAGES USING DCT

Abstract— The Structural information in an image is useful in many image processing applications like automatic visual detection of shapes and objects, to find defects in texture images, object detection, detection of defects in fabric image, digital image inpainting. Edges preserve the structural information in an image while discarding other details. Many methods exist based on edge detectors to arrive at the structure of the given image but the presence of noise affects their performance. In this paper a new method for finding structural information of image blocks using DCT is proposed. The image is divided into blocks and two dimensional DCT is applied to each block. With the frequency coefficients of the image blocks the structural information of the image blocks is found out and whether the block is rich in vertical or horizontal components along with the contrastness of the image block is analyzed. A comparison between edge detectors and the proposed method is made. The experimental results show that the proposed method works better than existing edge detecting methods.

IMAGE RECOVERY IN A CONTACTLESS FINGERPRINT IMAGE

Abstract—The conventional methods of acquiring fingerprint through scanners have some disadvantages. The main disadvantage is that the scanning device is not portable and therefore it can be used in the place only where it is installed. The method of acquiring fingerprint through contactless technique overcomes this difficulty since the fingerprint is captured using camera. This method has some disadvantages; some regions may be missing through acquisition. These missing regions are filled using the filling techniques discussed here.

TEXEL IDENTIFICATION USING K-MEANS CLUSTERING METHOD

Abstract -  Identifying the smallest portion of the image that represents the entire image is a basic need for its efficient storage. Texture can be defined as a pattern that is repeated in a specific manner. The basic pattern that is repeated is called as Texel(Texture Element). This paper describes a method of extracting a Texel from the given textured image using K means clustering algorithm and validating it with the entire image.  The number of gray levels in an image is reduced using a linear transformation function. The image is then divided in to sub windows of certain size. These sub windows are clustered together using K-means algorithm. Finally a heuristic algorithm is applied on the cluster labels to identify the Texel, which results in more than one candidate for Texel. The best among them is then chosen based on its similarity with the overall image.  The similarity between the Texel and the image is calculated based on then Normalized Gray level co-occurrence matrix in the maximum gradient direction. Experiments are conducted on various texture images for various block sizes and the results are summarized

COMPUTER VISION BASED APPROACH FOR INDIAN SIGN LANGUAGE

CHARACTER RECOGNITION

Abstract- Deaf and dumb people communicate among themselves using sign languages, but they find it difficult to expose themselves to the outside world. This paper proposes a method to convert the Indian Sign Language (ISL) hand gestures into appropriate text message. In this paper the hand gestures corresponding to ISL English alphabets are captured through a webcam. In the captured frames the hand is segmented and the state of fingers is used to recognize the alphabet. The features such as angle made between fingers, number of fingers that are fully opened, fully closed or semi closed and identification of each finger are used for recognition. Experimentation done for single hand alphabets and the results are summarised .

INDIAN SIGN LANGUAGE GESTURE SEGMENTATION USING ACTIVE CONTOUR

Abstract – Deaf and dumb people communicate using sign language which everyone can’t understand. The sign language can be conveyed to others using computers. This paper proposes a segmentation method using active contour algorithm that identifies the hands, which can be converted to medium so that everyone can understand. The gestures for Indian language are segmented and the contour is formed based on local information. The contour is the outline or boundary of an image. The local contour can segment objects with heterogeneous features. First the contour based segmentation is compared with edge based segmentation. Then the study of the energy in response to changes in localization is observed. And then the results of overlapping images that are possible with contour based segmentation are illustrates. This proposed method gives the exact contour of every gesture.

INDIAN SIGN LANGUAGE CHARACTER RECOGNITION USING NEURAL NETWORKS

Abstract- Deaf and dumb people communicate among themselves using sign languages, but they find it difficult to expose themselves to the outside world. This paper proposes a method to convert the Indian Sign Language (ISL) hand gestures into appropriate text message. In this paper the hand gestures corresponding to ISL English alphabets are captured through a webcam. In the captured frames the hand is segmented and the neural network is used to recognize the alphabet. The features such as angle made between fingers, number of fingers that are fully opened, fully closed or semi closed and identification of each finger are used as input to the neural network. Experimentation done for single hand alphabets and the results are summarized. 

DYNAMIC INDIAN SIGN LANGUAGE CHARACTER RECOGNITION:

USING HOG DESCRIPTOR AND A CUSTOMIZED NEURAL NETWORK

Abstract- Communication is the exchange of thoughts, messages, or information, by speech, visual signals, writing, or behaviour. Deaf and dumb people communicate among themselves using sign languages, but they find it difficult to expose themselves to the outside world. This paper proposes a method for recognizing Indian sign language (ISL) Two hand characters given as input by the user in the form of hand gestures. There are 18 ISL characters which are shown using two hands. Unlike the conventional method, this method does not require any additional hardware and makes the user comfortable. The system takes input at real time through a webcam integrated in the laptop. The hand region is separated out from the background using skin segmentation and motion segmentation. Since the segregation of overlapping hands is difficult, two hand characters are identified using HOG features. A back propagation neural network is used as the learning algorithm to make the system adaptable for different users. The system has been tested for several people of varying skin complexions, in several environments and was found to have accuracy of about 90%. The accuracy mainly dropped due to the illumination of the environment and occlusion of hands involved in two hand gestures.

SEGMENTATION, TRACKING AND FEATURE EXTRACTION FOR INDIAN SIGN LANGUAGE RECOGNITION

Abstract-Sign Language is a means of communication between audibly challenged people. To provide an interface between the audibly challenged community and the rest of the world we need Sign Language translators. A sign language recognition system computerizes the work of a sign language translator. Every Sign Language Recognition (SLR) System is trained to recognize specific sets of signs and they correspondingly output the sign in the required format. These SLR systems are built with powerful image processing techniques. The sign language recognition systems are capable of recognizing a specific set of signing gestures and output the corresponding text/audio. Most of these systems involve the techniques of detection, segmentation, tracking, gesture recognition and classification. This paper proposes a design for a SLR System.

VIDEO BASED INDIAN SIGN LANGUAGE RECOGNITION SYSTEM FOR SINGLE AND DOUBLE HANDED GESTURES WITH UNIQUE MOTION TRACE AS FEATURE

Abstract-Gestures or Signs are the means through which audibly challenged people communicate  with each other. Sign Language Recognition Systems computerize the job of a sign language translator. Our SLR system is built using powerful Image processing techniques. The system architecture involves techniques for skin color segmentation, Motion-Trace feature extraction and DTW Classification. Our proposed system takes as input a video of the gesture and outputs the text corresponding to the gesture performed. Gestures and signs are used interchangeably used throughout the paper. The system has an overall accuracy of 74.14%

CONVERSION OF BRAILLE TO TEXT IN ENGLISH, HINDI AND TAMIL LANGUAGES

Abstract- The Braille system has been used by the visually impaired for reading and writing. Due to limited availability of the Braille text books an efficient usage of the books becomes a necessity. This paper proposes a method to convert a scanned Braille document to text which can be read out to many through the computer. The Braille documents are pre processed to enhance the dots and reduce the noise. The Braille cells are segmented and the dots from each cell is extracted and converted in to a number sequence. These are mapped to the appropriate alphabets of the language. The converted text is spoken out through a speech synthesizer. The paper also provides a mechanism to type the Braille characters through the number pad of the keyboard. The typed Braille character is mapped to the alphabet and spoken out. The Braille cell has a standard representation but the mapping differs for each language. In this paper mapping of English, Hindi and Tamil are considered.

ECONOMIC PRINTING OF BRAILLE DOCUMENTS

Abstract- The objective of the paper is to explain the process of printing Braille documents using a dot matrix printer. Braille language has been the only medium of communication for the blind without the help of other people. Hence this project was attempted to help them and create an efficient and economical solution for the same. The Braille documents were printed using a dot matrix printer after removing the ink ribbon. Due consideration was given to the standard braille size and the accuracy of the impression made on paper. Various trials were done by changing paper quality and the best out of them was chosen.

STUDY ON BRAILLE INPUT OUTPUT DEVICES

Abstract- This document is about the Braille devices .there nearly 45 to 50 million are many people in the world who are blind and more than 269 who visually impaired. before the inversion of Braille device blind people cannot able to read or gain knowledge through reading Louis Braille is the person who invented the device called Braille which became popular by the usage ,nowadays there are many devices arrived in this world to solve the visually impaired peoples problem not only we can use such devices but also we can reduce the cost by combining various techniques. This paper is about the various Braille devices and their methods or their combination to make them cost effective.

AUDIO STEGANOGRAPHY

Abstract- Security over any communication medium has been an important concern. ‘Steganography’ - the act of hiding or secret writing is emerging to counter the attacks. A generic steganographic process is embedding (hiding) data in cover media and producing stego-media. This paper deals with LSB Insertion method and Bit Modification Methods. Drastic and obvious disturbance may occur during the steganography process and the stego medium may reveal the existence of the hidden data, which is what an attacker may be looking for. This paper focuses on how these problems can be overcome by using a proposed method and analyses the performance of these methods for hiding text, audio in audio media.

FABRIC DEFECT DETECTION USING GRAYLEVEL CO-OCCURRENCE MATRICES (GLCM)

Abstract-The task of fabric defect detection is carried out by human visual inspection in most of the traditional textile industry. The possibility of automated defect detection using gray level co-occurrence matrix is investigated in this paper. The method is based on calculation of Gray level co-occurrence matrix and extraction of texture features from it. Detection of defects is achieved by partitioning the image into nonoverlapping subwindows and classifying each subwindow as defective or non-defective with a Mahalanobis distance classifier being trained on defect free samples apriori. The experimental results demonstrating the use of this method are also presented.

FABRIC DEFECT DETECTION USING STATISTICAL TEXTURE ANALYSIS

COMPARISON OF GLCM AND HISTOGRAM.

Abstract- Quality assessment is an important task in textile fabric industry. In this paper defects in the fabric are identified through image processing techniques. The fabric is considered as a textured image and statistical texture analysis methods like Grey level co-occurrence matrix (GLCM) and Histogram are used for identifying the defects. This paper presents the performance analysis of GLCM and Histogram in identifying the defects. Commonly occurring 16 fabric defect images are used for experimentation. The aim is to choose the appropriate feature and technique for Fabric Automatic Visual Inspection system such that the discrimination between normal and defective regions is high. The performance of various features in defect detection is analysed and tabulated.

LOCATING FABRIC DEFECTS USING GABOR FILTERS

Abstract- Quality inspection is the key aspect in a fabric industry. Developing an automatic visual inspection system requires a robust and efficient algorithm for finding defective area. Locating the defective area is required to cut the cloth at appropriate locations. In this paper Gabor filters are used to find the appropriate location of defects in a fabric image. The parameters of the filter bank are tuned before using it for the location identification. This method is tested for 16 different defects that commonly occur in fabric industry and the results are summarized.

FACE TRACKING IN VIDEO BY USING KALMAN FILTER

Abstract-  Face Tracking has been one of the most studied topics in computer vision literature. Facial feature extraction has some problems which must be researched. Small variations of face size and orientation can affect the result of face tracking. Since the input image is captured from a surveillance camera, certain conditions have to be considered - like different levels of brightness, shadows and clearness - which are challenges for detection and tracking purpose. Most facial feature extraction methods are sensitive to various non-ideal such as variation in illumination, noise, orientation, time-consumption and color space used. So there is a need for a good feature extraction method that will enhance the quality and performance of face recognition system. First, segmentation of foreground and background object is the one by using histogram equalization. By this method we are able to segment face based on skin color. After segmenting, Kalman filter is used to track the faces under several conditions. This feature is helpful for the development of a real-time visual tracking control system.

VEHICLE DETECTION IN STATIC IMAGES USING COLOR AND CORNER MAP

Abstract— this paper presents an approach to identify the vehicle in the static images using color and corner map. The detection of vehicles in a traffic scene can address wide range of traffic problems. Here an attempt has been made to reduce the search time to find the possible vehicle candidates thereby reducing the computation time without a full search. A color transformation is used to project all the colors of input pixels to a new feature space such that vehicle pixels can be easily distinguished from non-vehicle ones. Bayesian classifier is adopted for verifying the vehicle pixels from the background. Corner map is used for removing the false detections and to verify the vehicle candidates.

NIGHT TIME VEHICLE DETECTION FOR REAL TIME TRAFFIC MONITORING SYSTEMS: A REVIEW

Abstract- Real time traffic surveillance using computer vision system is an emerging research area. Many new algorithms are being developed to perform the surveillance in the most effective manner. The first and critical step in these road traffic monitoring systems is to detect and track the vehicles. In this paper, we provide a brief review on the night time vehicle detection techniques that have been used in the recent years. The detection of vehicles in the night time can prove to be challenging because the usual features of the vehicles like the vehicle shadows, horizontal and vertical edges that helps in the identification in day time cannot be used during the night time. The only salient features that are visible in the night time are headlights, rear-lights and their beams, street-lamps, horizontal signals such as zebra crossings and traffic scenes with reflectors. Thus, in night time surveillance the target objects are the vehicle headlights and rear lights.

A SURVEY ON ALGORITHMS OF SHADOW REMOVAL IN VEHICLE DETECTION

Abstract - A shadow appears on an area when the light from a source cannot reach the area due to obstruction by an object. Shadow detection and removal in real time images is always a challenging task. In contrast with the rapidly expanding and continuous interests on this area, Shadow detection is critical for robust and reliable vision-based systems for traffic flow analysis. In our case Vehicle Detection, Shadow Removal is used to prevent moving shadows being misclassified as moving objects (or parts of them), thus avoiding the merging of two or more objects into one. The environment considered is Local Highway with multiple lanes and a fixed camera. This paper is aimed at giving comprehensive algorithms to detect and remove vehicle shadows in Real-time Videos.

A SURVEY ON VISION BASED FIRE DETECTION IN VIDEOS

Abstract- Computer vision techniques are largely used now a days to detect the fire. There are also many challenges in judging whether the region detected as fire is actually a fire this is perhaps mainly because the color of fire can range from red yellow to almost white. So fire region cannot be detected only by a single feature color many other features have to be taken into consideration. This paper is a study of the recent techniques and features extracted by different existing algorithms .

PYRAMID–BASED IMAGE INTERPOLATION

Abstract— A technique for better reconstruction of a distorted region in an image using Pyramid-based image interpolation is presented in this paper. The central idea is to construct the image pyramid, so that the apex of the pyramid has a minimized distorted region. The reconstruction of the original image starts from the apex of the pyramid. Each time, the pixels in the distorted region are recovered using bilinear interpolation technique. Since the distorted pixels in the image at the apex of the pyramid have maximal undistorted neighbors, the accuracy of reconstruction is enhanced. Thus the bottom-up approach proposed by this paper results in improved image reconstruction when compared to top-down technique due to the above mentioned higher availability of undistorted pixels in the neighborhood of a distorted pixel.

A SURVEY OF DIFFERENT STAGES FOR MONITORING TRAFFIC RULE VIOLATION

Abstract: A traffic surveillance system is a controlled system th at helps to monitor and regulate the traffic. In this paper, a method for extracting the license number of the vehicle that is exceeding the speed limit is proposed. A Study is conducted by covering various stages of monitoring system such as vehicle detection in the video, tracking the vehicle for speed calculation and extracting the vehicle number in the number plate that can be used in places with high public vicinity.

SURVEY ON TRACKING ALGORITHMS

Abstract—Tracking non-rigid bodies like hands and face are of great significance in Sign Language Recognition Systems. (SLR Systems).In the gesture and sign language video sequences, the movement of the hand and face tends to be rapid and involves a lot of occluding scenarios. Thus we understand that naïve color based tracking methodologies are clearly insufficient. To improve the performance of tracking system a predict-update framework may be employed but this requires careful initialization of parameters, since the tracker tends to drift and lose track of these objects in the required sequences. This paper provides the survey of the various methods that have been used for determining the trajectory of the objects.

SURVEY ON VEHICLE DETECTION TECHNIQUES

Abstract—Vehicle detection is the foremost step in monitoring the speeding vehicles in a highway. The video sequences captured by a stationary camera show us that there’s a need for a vehicle detection algorithm which handles sudden illumination change and also the scenarios where the foreground merges into the background. This paper provides us a survey of various background subtraction techniques that are used for detecting the vehicles efficiently.

SURVEY ON HAND GESTURE RECOGNITION

Abstract- The main goal of gesture recognition is to create a system which can recognize specific human gestures and use them to convey information. Types of techniques used in recognizing hand postures and gestures are compared along with advantages and disadvantages of each. Gestures considered as a natural way of communication among human, since it is a physical movement of hands, arms, or body which conveying meaningful information and helps in expressing thoughts and feelings effectively.

SURVEY ON SKIN COLOUR SEGMENTATION TECHNIQUES

Abstract— Segmentation of non-rigid complex structures is of great significance in the field of Vision Based Applications. There are many techniques available to segment the non-rigid bodies like hand, face, etc. These techniques are very useful in applications of Human Computer Interaction, Virtualization, Sign Language Recognition Systems, etc. The results from segmentation are used to identify the regions that are further processed and analysed to fit their application needs. Segmentation techniques have been broadly classified into two types – Parametric & Non – Parametric. The aim of this paper is to provide a survey of the various techniques used for parametric pixel based skin colour segmentation. The paper also provides the understanding of the segmentation techniques that are particularly most suited in a particular scenario over the other techniques.

LITERATURE SURVEY ON GESTURE CLASSIFICATION TECHNIQUES

Abstract— Gesture Recognition is been a prominent method in making human-computer interaction system. One of the main application of any gesture recognizer is sign language recognition. Particularly in this field a lots of advancements has been bought such as from identifying a static or isolated action to identifying continuous or dynamic gestures. The process of recognition of dynamic gestures involves various steps such as segmentation of ROI, tracking of key point, feature extraction and classification of gesture. In the process of gesture recognition, classification which is the final step involves computerized processing of the data which has been acquired from the actions or gestures performed and determine whether the data corresponds to a particular gesture. For improving the accuracy of recognition, various pattern recognition or machine learning algorithms as HMM, Artificial Neural Networks, and fast DTW. The main purpose of this paper is to analyze these methods and compare them, enabling the reader to find an optimal solution for their problem.