Identification of Volcano Hotspots by using Resilient Back Propagation (RBP) Algorithm Via Satellite Images
Data Hiding in Encrypted Compressed Videos for Privacy Information Protection
Improved Video Watermarking using Discrete Cosine Transform
Contrast Enhancement based Brain Tumour MRI Image Segmentation and Detection with Low Power Consumption
Denoising of Images by Wavelets and Contourlets using Bi-Shrink Filter
In this research work, an algorithm has been developed for robust image processing using Discrete Cosine Transform (DCT) in Fuzzy Inference System (FIS) system. By altering the watermarking coefficient, a value is ascertained which has a better PSNR value leading to the development of a robust algorithm. The same algorithm can be applied universally to various images, whether they are in colour or black and white. The sensitivity of the perception of the eye towards light and colours in an image are narrowed down to three fields namely, luminescence, edge and contrast, which are used as inputs to the Mamdani type FIS subjected to a knowledge base. The watermark created is inserted over the weighted output received from the FIS output. High value of PSNR and low MSE brings out that the algorithm used for inserting the watermark is robust in nature. Using the Cox's algorithm, the watermark is taken out from the watermarked image. The watermarked image has been successfully transmitted over the VHF radio set held with the Indian Army. Due to the implementation of DCT for image processing, the original image has been compressed to a few kilobyte size for it to be successfully transmitted over the radio frequency.
Due to the refraction, absorption, and scattering of light by suspended particles in water, underwater images have low contrast, blurred details, and colour distortion. The Very-Deep Super-Resolution (VDSR) reconstruction model is introduced to increase the resolution of images captured using underwater applications. A residual learning model for underwater image enhancement has been introduced in this paper. The CNN layers are estimated and applied to the images in order to obtain the feature map. Based on this feature map, the particles are removed. According to the underwater image enhancement experiments and a comparative analysis, the colour correction and detail enhancement performance of the proposed methods are superior to that of previous deep learning models and traditional methods. The experimental results suggest that this method produce better results when compared to state-of-art methods.
The traffic lights are pre-programmed to wait for a certain fixed time after every change in signal. The operation of the traffic lights does not depend on the traffic on the roads and remains constant during its operation. A simple way to ease road traffic is by allocating more time for the vehicles to pass on from heavy traffic roads than roads with relatively less traffic. The proposed system in this paper uses a camera to capture the images of the moving vehicles that connects the traffic signal. The pictures captured are then processed to determine the total number of vehicles present on each road at that instant. The system uses image acquisition, image scaling, image enhancement, followed by object detection in order to estimate the total number of vehicles on roads to regulate the traffic. The dynamic time allocation for traffic on each road is determined based on the actual traffic on the road, and this system will control traffic more effectively than a predetermined time for each road. The system is cost-efficient and does not require any installation of complex machinery to monitor the complete traffic.
Real time lip localization is a significant research topic in the computer vision area as it has a greater number of applications such as biomedical image analysis, human computer interfaces, automatic speech recognition, speaker detection and lip-reading system. The most popular face detection algorithm is Viola and Jones based on statistical methods. This paper explains a technique used to localize lips from the face image. In first scenario, mouth detection is done using Viola-Jones algorithm. In second scenario, mouth region is detected using hybrid method, which is a combination of Viola-Jones algorithm and geometric projection method.
Fire hazards result in huge damage to the natural resources, public properties, and loss of human and animals lives. Huge fire catches up in remote areas and expands suddenly to nearby areas and eventually becomes uncontrollable. Though sensor based fire detectors are available in market, its function is dependent on the proximity of the sensor to the fire incident, and may not be useful for detecting fire in a remote place where a fire incident is not expected. Hence electronic fire detection systems are inefficient by its nature of operation. Therefore, this article proposes a method for effective fire detection, regardless of the distance of the fire sensing system. The fire detection mechanism is based on the motion occurring between the blocks of a reference frame and the current frame in the test video. This motion estimation based method is capable of detecting the presence of fire at a distance that is better than fire detection systems based on fire sensors. Fire detection performance analysis of the system is found varying based on the block size chosen in motion estimation algorithm. Similarly, detection performance analysis had been found based on the distance of the fire measured from location of the camera. The results obtained in this paper show that the proposed algorithm is suitable for fire detection with a moderate coverage distance with scope for future improvement.