Anil K. Roy

Enhancing effectiveness of motor rehabilitation using kinect motion sensing technology

Serious games and their use in healthcare applications are rapidly growing areas of interest for researchers and practitioners. It has been observed that the physically disabled persons typically lack enthusiasm for rehabilitation. This hinders their recovery and ability to care for themselves. Using research and recommendations from therapists, we developed a product, Kinect-o-Therapy, that involved implementation of exercise routines targeting different parts of the body and aided people with motor disabilities stemming from cerebral palsy, spinal cord injury, post stroke and hereditary muscle ailments amongst others. Kinect-o-Therapy incorporates serious games that process 3D input using the Microsoft Kinect motion sensor to build an effective rehabilitation system. It is a plug-and-play type product that gives feedback to the patient and the doctors to monitor performance. The game environment and feedback mechanism ensure patients involvement and high motivation to carry out these exercise routines regularly. Real-life testing of Kinect-o-Therapy shows that patients and physiotherapists both find it quite engaging, easy-to-use and a motivational alternative of rehabilitation.

A Morphology Based Approach for Car License Plate Extraction

Locating the car license plate in an image or video frame of a car is an important step in car license plate recognition/identification applications. This problem poses many challenges like location of license plate from images taken in poor illumination and bad weather condition; plates that are partly obscured by dirt and images that have low contrast. This paper presents a new morphology based method for license plate extraction from car images. The algorithm uses morphological operations on the preprocessed, edge images of the vehicles. Characteristic features such as license plate width and height, character height and spacing are considered for defining structural elements for morphological operations. Connected component analysis is used to select the band containing license plate from the candidate segmented. The experimental results with a reasonably large set of car images are very encouraging.

Exposing structural tampering in digital images

Manipulation in a digital image may appear because of unintentional or harmless post-processing of the image, such as through JPEG compression or contrast enhancement. But some manipulations or tampering in a digital image are done with malicious motivation. One may find standard techniques to identify and analyze many types of such manipulations. But with the phenomenal growth in processing power of image capturing devices and user-friendly complex post-processing software, it has become a real challenge to find a technique that could tell exactly where, on the image, the tampering has been done. We propose a new method that could identify the block where structural tampering has been done. This method automatically ignores the impact of harmless changes in the digital image. Our proposal is based on the unique feature of each image that is described by the hash function. We find the correlation coefficients of the original image and the tampered image with respect to a reference image and establish that the difference of these two correlation coefficients is significantly larger, hence exposing the tampering. The method is tested on a database of images conforming to our theoretical analysis.

Digital image tampering detection and localization using singular value decomposition technique

Recent years have witnessed an exponential growth in the use of digital images due to development of high quality digital cameras and multimedia technology. Easy availability of image editing software has made digital image processing very popular. Ready to use software are available on internet which can be easily used to manipulate the images. In such an environment, the integrity of the image can not be taken for granted. Malicious tampering has serious implication for legal documents, copyright issues and forensic cases. Researchers have come forward with large number of methods to detect image tampering. The proposed method is based on hash generation technique using singular value decomposition. Design of an efficient hash vector as proposed will help in detection and localization of image tampering. The proposed method shows that it is robust against content preserving manipulation but extremely sensitive to even very minute structural tampering.

Non-blind method of detection and localization of structural tampering using robust Hash-like function and similarity metric for digital images

Content preserving manipulations (CPM) in a digital image has been discussed subjectively in the literature so far. A number of ways of obtaining robust hash functions have been reported which claim to be robust against any content preserving manipulations and sensitive to structural tamperings. However, in many cases, one can make out these structural tampering just by viewing and comparing the two images, the original and the tampered image. We report an algorithm using robust Hash-like function and similarity metric by which even a very small but malicious tampering affecting only a few pixels of the digial image could be localized accurately. Results are shown in which locations of single or even multiple structural tamperings are successfully identified.

A novel method for detecting light source for digital images forensic

Manipulation in image has been in practice since centuries. These manipulated images are intended to alter facts — facts of ethics, morality, politics, sex, celebrity or chaos. Image forensic science is used to detect these manipulations in a digital image. There are several standard ways to analyze an image for manipulation. Each one has some limitation. Also very rarely any method tried to capitalize on the way image was taken by the camera. We propose a new method that is based on light and its shade as light and shade are the fundamental input resources that may carry all the information of the image. The proposed method measures the direction of light source and uses the light based technique for identification of any intentional partial manipulation in the said digital image. The method is tested for known manipulated images to correctly identify the light sources. The light source of an image is measured in terms of angle. The experimental results show the robustness of the methodology.

ImPatho - an image processing based pathological decision support system for disease identification and a novel tool for overall health governance

Governance of healthcare has been one of most challenging and complex issues of the government of any developing country. Many of the life threatening diseases, such as Sickle Cell Anemia, can be prevented properly if these can be diagnosed on time. In many cases, diagnosis demands a simple blood test to find out deformation in the shape of blood cells. We investigated in this paper, if such tests can be carried out at the doorstep of the likely patient wherein the intervention of a skilled person from the pathology lab is suitably replaced by a technology developed in form of an app. The app and the hardware combo which we propose, called ImPatho, uses image processing algorithm to judge if blood cells are indicative of some disease. Our solution then follows a decision support system through data analyses which can be reported to doctors, district health officers, state health department for proper treatment and monitoring, via our app. It is shown that the developed system, ImPatho, is a complete end-to-end solution for the challenges faced by the government of developing countries particularly where people from remote, tough or even rural terrain cannot come to a nearby pathology lab to get themselves investigated for a likely disease. Our solution ImPatho empathizes this Catch 22 situation and makes an attempt to bring the pathology lab at the doorstep of such helpless people. This is how our system helps ensure proactive, responsive healthcare governance. ImPatho has a huge potential, if scaled to higher levels of governing echelon, to provide an alternative solution for governments who are sensitive towards healthcare of poor people.

Comprehensive image index and detection of tampering in a digital image

Integrity of digital image is of prime importance in areas relating to police investigation, evidence in the court, medical science and biometric images. With easy availability of image editing software and large volume of digital images in daily life, detection of image tampering has become a serious challenge for image processing community. Image content based hash function for authenticating an image is being widely used. In this paper we define Comprehensive Image Index that tells about the type of tampering done in a digital image. Most of the existing hashing technique extract one feature of the image and develop a single index hash function for image tampering detection. Proposed technique uses a new 3-tupled comprehensive image index which incorporates indices corresponding to structural tampering, brightness level adjustment and contrast manipulation. This technique is found to be robust against content preserving manipulations at the same time it is sensitive to malicious tampering.

Detection of structural tampering in a digital image using Canny Edge Detector

The verification of authenticity of digital images has become a challenging problem due to easy availability of image editing software which can also be used to tamper the images. Content based digital signature is being used by researchers for detecting malicious tampering operations. In the proposed scheme, Canny Edge Detector has been used to extract the feature vectors of the image through edge detection. This information is then used to generate a hash vector for tampering detection and localization. The suitable mathematical index has been developed to quantify the amount of tampering using the hash generated from digital content of the image.

Low power, high bit rate CMOS driver for VCSEL used in free space optical interconnect

We propose a novel driver circuit design using 0.8 micron CMOS technology that drives a 1550 nm VCSEL used in free space optics. We report a distinct improvement in power consumption at 2.5 Gbps.