Multi-Lingual Character Recognition and Extraction using Recurrent Neural Networks
A Novel Approach to Smart Autonomous Monitoring of Indoor Plant Health Based on Leaf Color
Edge Preserving Image Denoising with Deep Networks and Ensemble Convolutional Neural Network Architectures
Neural Network Approach to Detect Renal Calculi
Enhancing Fabric Quality Control: Implementing Real-Time Defect Detection with Image Processing Techniques and Arduino
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
Receiving videos with least error has become one of the main challenges in video applications in recent years. Video error occurrence may be because of loss of data. Video error concealment is technique in which error is removed from the data. In video error concealment, data is recovered back completely. The technique used for that is moment invariance. Video applications are widely used in many domains, such as internet videos, entertainment media, and security applications. In this paper, Video Error Concealment is achieved by Moment Invariance using MATLAB. Some standard videos with different errors in different frames are used to analyze the quality of videos. Error is detected using error function moment invariance technique and corrected by using block matching algorithm. In Moment Invariance method, frames are divided into macroblocks. For analysis, error in macroblocks of different sizes like 16*16 pixel, 8*8 or 4*4 pixel macroblocks are considered in this paper. Error is introduced with different sizes of blocks until error concealment is done. The quality of the video is analyzed using Structural Similarity Index Measurement (SSIM) and Peak Signal to Noise Ratio (PSNR). Result indicates that quality is improved for different error videos.
Analysis of clouds and their physical properties, such as liquid water content, ice water content, and Reflectivity plays a crucial role in examining the precipitation rate. Till now much work has been done on the NOAA data to examine the clouds and their relation to the rainfall rate. In present work, clouds are detected and classified based on the NOAA-18 AVHRR (Advance Very High Resolution Radiometer) satellite imagery using the SVD (Singular Value Decomposition) property. The eigen values in the SVD help to distinguish between land, snow, and ocean based on the spectral features of the NOAA band 1, 2, and 4 images. The proposed method found the detected clouds with accuracy of 60% using statistical measures. The RGB satellite images are extracted from the NOAA-18 data using ERDAS imagine software which are useful for further processing using MATLAB. All the data used in this work are acquired from NOAA-18 AVHRR satellite imagery installed at S.V. University College of Engineering.
Motion estimation is a fundamental procedure for video compression. It is directly related to the compression efficiency by reducing temporal redundancies. Motion estimation is the most critical part of a video encoder and 50% coding complexity or computational time depends on it. To minimize the computational time, there were various ME algorithms proposed and implemented. In this paper, the authors provide performance analysis of Star refinement on Diamond search algorithm and after evaluation, they determine the most optimal algorithm. Each algorithm is evaluated using many test videos and compared through Peak Signal to Noise Ratio (PSNR) and per macro block search points (i.e. computation time) along with search areas. Results suggest that among all the evaluated algorithms, Star Diamond- Diamond Search has the best PSNR based on computation time.
Rice is the staple food for major population of India. Area and spatial distribution information of paddy rice are important for understanding of food security, water use, greenhouse gas emission, and disease transmission. Due to urbanization and drought during various times, rice cultivation is negatively affected in Karnataka. An algorithm based of paddy phenology is used to identify paddy fields and map the same. Landsat 8 data with high temporal resolution and geographic coverage is used for mapping. Envi along with idl 8.3 is used for data processing. The resultant paddy rice map well supports the statistical data. The resultant paddy rice map is expected to provide unprecedented details about the area and spatial distribution of paddy rice fields in Koppal, which will contribute to food security assessment, water resource management, estimation of greenhouse gas emissions, and disease control if this work is extended towards more scenes in a sub continental level.
The farmers are confronted with new difficulties consistently. The sufficient usage of water, disease associated with crops; trespassing, etc, are majorly faced problems by farmers in the current world. Total 7.68% of Global Gross Domestic Product (GDP) is accounting from agriculture. Computers have been utilized to motorization and computerization and to build up a choice emotionally supportive network for taking key choices on the agrarian generation and insurance look into. It is vital to design an intelligent computer vision algorithm to minimize the risks in farming. As plant disease estimation and growth is still carried out due to the visual nature of the plant monitoring task, computer vision techniques seems to be well adapted. Early identification of disease in plants helps in avoiding pests and to take countermeasures. Diseases are analyzed by different digital image processing techniques.
