Search

Medical imaging has grown tremendously over years. The CT and MRI are used at most. MRI is less harmful, but one cannot underestimate the side effects of CT. Current study demonstrate the serious risk of cancer, due to repeated use of harmful X-rays. Consequently the plan of low dose imaging protocol is emerging as vital solution in the present scenario. Low dose imaging is achieved by sparse view scanning. The sparse-view x-ray computed tomography (CT) is appreciated in medical diagnosis and industrial non-destructive testing. However, when the sampling is insufficient due to sparse view, the reconstructed image usually suffers from complex artifacts and noise. In order to deal such issue, we present interpolation technique to fill the sinogram as promising method to address issues of low dose imaging. Reconstructed image quality parameters are recorded, and are compared to original image and the low dose image. The evidenced results convey substantial improved performance compared to conventional low does image.

The important role played by the data compression is in reducing the storage space required, and ensuring bandwidth utilization in efficient manner and low transmission delays. Image compression is also a part of data compression. Images contain important data that is to be transmitted or stored. The compression must not degrade the quality of the image, and should be acceptable. This paper presents simple, low complexity image compression algorithm. Algorithm scans the image in a raster san manner in terms of 4X4 blocks. Compression is achieved by the diagonally situated pixels and indices their surrounding pixels of the block. The mages produced by proposed algorithm are high fidelity images at low PSNR values. The results are comparable with standard algorithm results. The reconstructed image quality is better than JPEG images at same PSNR.

Images consume more space when stored and use a lot of bandwidth during transmission. Hence, an image compression reduces storage space required, and time to transmit an image by making effective usage of bandwidth. In this paper, a simple image compression method using an index-based approach is proposed. This approach makes use of simple indices of an array to compress images. The context of nearby fifteen pixels is used, and the index for the current pixel is found. The indices are found for every pixel of an image. These indices are transformed and written into a transformed file. This transformed file is further compressed with Huffman compression to achieve a compressed file. Decompression uses a simple methodology to reconstruct the image. The pixel value is reconstructed using the index, and from previously reconstructed pixels of the image. The proposed algorithm provides high fidelity images at lower Peak Signal to Noise Ratio (PSNR). The results obtained from the proposed algorithm are compared with JPEG algorithm. At the same PSNR values, the quality of the reconstructed images by the proposed algorithm is comparable to that of the JPEG algorithm.

Data compression is the way of reducing the number of bits required to store and efficiently utilize bandwidth during the transmission of data over a network channel. Image compression is also having importance as images contain important binary data to represent as image and that is to be stored and transmitted. The compression of an image must not reduce the quality of the image except in some cases. This paper proposes a compression algorithm with low PSNR, high fidelity images. The algorithm alienates the input image as non-overlapping blocks of size 4*4. Compression is achieved by using diagonally situated pixels and other adjacent pixels of the blocks. The decompression uses a simple approach. The diagonally situated pixels and other adjacent pixels are used to reconstruct images in decompression phase. The proposed algorithm reconstructs images with high fidelity at lower Peak Signal to Noise Ratio (PSNR). The results of proposed algorithm are compared with JPEG. Results have shown that at same PSNR values, reconstructed image quality by the proposed algorithm is better than that of the JPEG algorithm.

Digital image storage and transmission is a challenging task nowadays. As per statistics, an average of 1.8 billion images are transmitted daily. Hence, image compression is inevitable. Here, we discuss a novel approach, which is a type of lossy image compression. But the quality of reconstructed image is higher. This method makes use of reducing the number of bytes by storing the number of ‘1’s in each bitplane of three adjacent pixels and recording the count of number of ‘1’s. This count value is stored, and later Huffman compression is applied. Reconstruction is done by energy distribution method. This gives better results in terms of quality at lower PSNR values compared to JPEG algorithm. Here we have used another metric to measure the quality of the image which is discussed in detail.

In recent years, video surveillance systems evolve a great interest as the application area. Recent studies have proven integration of digital image processing, computer vision, artificial intelligence and data analytics into video surveillance application. This paper provides a systematic review of the state-of-the-art video analysis techniques available in machine learning and deep learning. Video processing research trends illustrate a survey on practices like object detection, object recognition, object tracking, traffic control and monitoring, action and behaviour recognition, disaster management and so on. We further present focus on computational approaches for various challenges of video processing techniques, objectives of those techniques and evaluation based on multiple problems. In the end, we have discussed emerging issues in the field of video analytics and how machine learning and deep learning contribute to those issues.

Digital image storage and transmission plays a very important role in today’s modern world, as most of the data transfer involves images. Hence, digital image compression is of great importance. The compression leads to either lossy or lossless type of images. Here, we discuss about a unique approach for lossy image compression, which involves a threshold. The number of pixels whose sum is lesser than a threshold is counted, and this count is saved in a file instead of actual pixel intensity values. A difference between the threshold and the computed sum is also stored. Later, reconstruction is done by reading the count and difference values. Average is calculated, and the count number of pixels is replaced with this value. We find that we can achieve better quality at lower PSNR values with this approach as compared to JPEG algorithm.

Population explosion leads to an unprecedented increase in the number of physical objects or vehicles on road. As a result, the number of road accidents increases due to a very heavy traffic flow. In this paper, traffic flow is monitored by using computer vision paradigm, where images or sequence of images provides a betterment on the road view. In order to detect vehicles, monitor and estimate traffic flow using low cost electronic devices, this research work utilizes camera module of raspberry pi along with Raspberry Pi 3. It also aims to develop a remote access using raspberry-pi to detect, track and count vehicles only when some variations occur in the monitored area. The proposed system captures video stream like vehicles in the monitored area to compute the information and transfer the compressed video stream for providing video based solution that is mainly implemented in Open CV by Python Programming. The proposed method is considered as an economical solution for industries in which cost-effective solutions are developed for traffic management.

The term Internet of Things [IoT] refers to the ever expanding complex system of basic things that emphasize communication between computing objects, devices and systems by offering connectivity from anyplace and at any time. It is estimated that by the end of 2020, 50 billion devices are said to be connected. IoT technologies play a crucial role to encompass many smart applications in real life. On the other hand, the crosscutting nature of IoT components and systems, introduce new security challenges. IoT covers advantages for various fields like agriculture, industry, healthcare, automobiles and home automation for improving and automating various day-to-day activities. Flood is usually caused either by change in the state of water body or due to the overflow of rivers, dams, etc. Due to advanced civilization and improved human life, environment problems also tend to increase. This paper includes the effective and flexible method for the detection of flood and alerting system. The most advanced technologies like machine learning (ML) provide significant boon to the field of technology are very powerful in monitoring the normal and abnormal behavioral characters of any machine. The objective of this work is to survey on flood issues. Neural networks are most popular, widely used for rainfall forecasting and perform efficiently.

Target/object detection, recognition, position, movement speed, etc. is easy when the object is near or easily visible. But, the same doesn’t stand true especially when the object is far or not visible due to so many factors like weather conditions, day/night cycle, etc. Therefore, Radio Detection And Ranging (RADAR) was invented, which uses radio waves to determine the range, angle, or velocity of objects. But, it uses long time to detect, has short detection range, not target specific because of wide range, oversensitive, costly, etc. A cheaper, easy and effective alternate solution is to use ultrasonic sensor which use sound waves for detection and ranging. Therefore, this paper provides a method in which the Ultrasonic Sensor (HC-SR04) acts as RADAR. The HC-RS04 is connected to Servo Motor (SG90) for the rotation/movement purpose. SIM808 module is also used to notify object detection via message/SMS. These components are connected to Arduino Uno and Raspberry Pi3 for being processed to detect and notify the object. Usually, the range of ultrasonic wave is 20kHz but here the HC-SR04 range is 3cm to 4m as it is smaller in terms of project usage. Advantages are: it is not affected by color or transparency of objects, can be used in dark environments, not highly affected by dust, dirt, or high-moisture environments, etc. The results show the object detected with its range/distance and angle in a java based GUI, different ranges of object in cm at which it is detected and the detection message sent to the admin.

Most of the institutional authorities are troubled with the cumbersome method of maintaining manual attendance of their employees. The manual process of signing on a paper is prolonged and insecure. An efficient attendance monitoring system needs to be enforced at such places. Radio Frequency Identification (RFID) based attendance system provides us with a solution that caters to issues like proxy attendance. This paper describes the design of an RFID based attendance monitoring system which uniquely identifies each employee/student based on their RFID tag which is attached to their ID card. This makes the mechanism of recording the attendance effortless, quicker and protected as compared to conventional method. This system is designed to be used at different educational institutions, corporate offices, government offices etc. The proposed system consist of both hardware and software components based on IoT Technology. The hardware component consists of RC522 RFID card reader and RFID tags/cards. The software component consists of the Web-based GUI for viewing the employee’s or student’s attendance, which is hosted on a web server and which stores the data in a database server. The employees or students just need to place their RFID card or tag on the reader and their attendance will be recorded for the day. Also, the attendance recorded will be more accurate as the system is synced with a real-time clock.

Cloud computing is a distributed platform that offers dynamic provisioning of on-demand resources to the users. The rapid growth of cloud computing is due to its adaptive nature that brings enormous benefits to the organizations. OpenStack is a cloud operating system used to build public and private clouds. OpenStack consists of a number of interconnected entities that are combined to govern the processing of compute, network and storage resources. Resource scheduling in OpenStack deals with provisioning of compute resources. Nova-scheduler is a component of OpenStack Compute service which is responsible for resource scheduling. The default nova-scheduler used in OpenStack is Filter Scheduler. The scheduling approach presently used is ineffective in terms of cost, resource usage and energy cost per server. To address these issues, we propose a SLA (Service Level Agreement)-aware resource scheduling strategy which provides better gain to both customers and cloud service providers. We carry out the experiments using real-time multinode private cloud setup. The results reveals the effectiveness of the designed resource scheduling model.

This paper proposes a new lossless image compression technique which makes use of simple mathematical equations along with Joint Photographic Expert Group Lossless (JPEG-LS) prediction technique to compress the image. Context of four pixels is used and the four pixels are varied in their signs which constitute a system of sixteen equations. Among the four context pixels, three pixels are actual pixels whereas the fourth pixel is the predicted value of the current pixel. Equations are solved to obtain the equation which result in minimum positive equation. Having obtained the minimum positive equation, it is solved by replacing the predicted pixel by the actual pixel value. Resulting absolute value is written in the image which results in transformed image and the sign of the result is written in a file. Both the transformed image and the sign file are Huffman encoded to reduce their size further. Compression ratio achieved for Lena image using the proposed compression technique is 1.6.

This paper presents a new lossless image compression algorithm which uses Joint Photographic Expert Group-Lossless (JPEG-LS) prediction technique and the concept of the seed number for compression. Image is processed in blocks of size 4*4. Pixels of the blocks are placed in a table of size 16*16. The span across which the pixels of each block are spread in the table is determined which forms a rectangle. Seed number for that block is computed. Using the JPEG-LS prediction technique in regular mode, the pixels are predicted and their position in the rectangle is found. Huffman encoded seed number, rectangle position and the positions of the pixels constitute the compressed files. Image is decompressed by Huffman decoding the files and reconstructing the pixels with the help of the pixel position, rectangle position and the seed number. This approach gives a new technique to compress the image, It is a simple approach and though the compression ratio is slightly lesser compared to JPEG-LS, it can be improved by Iterative approach.

In order to meet the increasing need of high quality images in the current trend, it is very important to manage image data so that they can be efficiently transmitted, received and stored without any loss of information. This motivates us to perform a research on the JPEG-LS, which is the current standard for lossless image compression. In this paper a comparative study of JPEG-LS algorithm is presented. 7 lossless techniques are proposed and compared with the performance of JPEG-LS. 5 of proposed techniques use the casual template used in JPEG-LS with fixed predictor and limited to work in regular mode of operation which involves prediction followed by entropy coding. Rest 2 techniques are independent of JPEG-LS and perform encoding based on position information of each pixel. Huffman coding is used for entropy coding of prediction errors and transformed images. Experimentation is done using 10 standard grayscale images of size 512∗512 and results show that the JPEG-LS performs better.

Software Defined Networking (SDN) is a new promising networking concept which has a centralized control over the network and separates the data and control planes. This new approach provides abstraction of lower-level functionality and allows the network administrators to initialize, control, change, and manage network behavior programmatically. The centralized control, being the major advantage of SDN can sometimes also be a major security threat. If the intruder succeeds in attacking the central controller, he would get access to the entire system. The controller is highly vulnerable to Distributed Denial of Service (DDoS) attacks which lead to exhaustion of the system resources which causes non-availability of the services given by the controller. It is critical to detect the attacks in the controller at earlier stage. Many algorithms and techniques have been discovered for this purpose. But less work has been done in the field of SDN networks. Using machine learning algorithms for classifying the connections into legitimate and illegitimate is one such solution. We use two machine learning algorithms namely, the Support Vector Machine (SVM) classifier and the Neural Network (NN) classifier to detect the suspicious and harmful connections.

Medical imaging has been identified as one of the fastest growing of all health care sectors. Despite the overwhelming benefits of computed tomography (CT) in diagnosis, there is concern over radiation induced cancer. Retaining diagnostically acceptable medical image quality with minimum radiation dose is the current need in the healthcare system. In this paper we have optimized WH-HYPR low dose image reconstruction algorithm using Taguchi method of experimental design and also studied the clinical permissible low dose range. The Wright Huang-Highly constrained back Projection (WH-HYPR) algorithm is considered as most promising low dose image reconstruction technique. This technique involves the formation of composite and weighted image whose product results in HYPR image with good temporal and spatial characteristics. Taguchi Method is a statistical approach used to fine tune the process influenting parameters more methodically and speedily [1]. Through the research work it is found that the algorithm can be optimized by considering a static constraint angle of 120° while the formation of the backprojection and also CT datasets containing around 15 to 20 time frames with least noise, maximum 20% structural and contrast variability generated a better quality HYPR image. An angle of 1° was found to be the threshold incremental angle for clinically acceptable low dose CT datasets used in WH-HYPR reconstruction. The proposed WH-HYPR algorithm allows better retention of image quality when compared with other reconstruction methods like FBP, AIDR3D and IR for undersampled data.

The Internet of Things (IoT) defines that objects are interconnected through wired and wireless networks without user intervention. The current IoT perform, sensing, actuating, data gathering, storing, and processing by connecting physical or virtual devices to the Internet. The application of IoT brings the higher efficiency and the more convenient life. Meanwhile, data collection will benefit from the ability to detect changes in the physical status of things. Large quantities of low cost, unattended wireless sensor nodes may be deployed to monitor a wide range of environments. Water is an essential resource and its management is a key issue. The proposed work is IOT based system for management of water distribution on a large campus. Depending on the level of the water, the submersible pump starts and stops automatically. When a level of water in an overhead tank is minimum then submersible pump automatically starts pumping and when it reaches to maximum level then submersible pump stops automatically. If the level of water reaches the maximum then intimate the user through SMS and upload detected water levels to the cloud. This system works automatically without any human intervention.

Determining the vehicle density running on the specific road is an important topic in Intelligent Transportation System. Reliable and robust vehicle detection from video sequence is an important problem in the application to control the traffic. The processing time of the vehicle detection system should meet the real time and should be free from false detection. The density of the vehicles defines the usage of road, traffic signals, etc. If the congestion is high road widening should be taken place or the vehicle should be diverted to the other way so that congestion cannot take place the next time. The algorithm proposed in the system detects vehicle and track them in real time. Detection of the vehicles is purely carried on color features of the vehicles. After detection of the vehicles they are tracked using enhanced Kalman filter with the data association. Cost matrix for data association plays an important role in allocating a centroid to each enhanced Kalman filter. So, that Kalman filter tracks the same centroid. The number of vehicles running in a video or in particular lane can be determined. Detection, tracking, convex hull, connected component labeling, color extraction.

Multimedia security has advanced tremendously over the decades due to the change in variety and volume of data. In the current security context, intelligent systems for multimedia security are very much in demand. Various applications such as biometric, e-commerce, medical imaging, forensics, aerospace, and defense require high-end data security systems. Conventional cryptography, watermarking, and steganography fall in short to provide security for high-resolution 2D/3D image and high-definition video. Persistent demand exists for designing new security algorithms for 3D graphics, animations, and HD videos. Traditional encryption method does not suffice the current need, as its securing ability is limited when it gets decoded. Steganography techniques are reported for securing text, audio, and video content, but observed to be few in number compared to image steganography techniques. Watermarking techniques for securing video content, text, and animations are reported in the literature but seem to be few in numbers as compared to image watermarking techniques. Majority of the literature is observed to apply digital watermarking as security means for video and image data. However, digital watermarking for 3D graphics is a current research topic. On the other hand, video watermarking techniques shall be broadly classified based on domain and human perception. Usually, video watermarking techniques do not alter video contents. But current trend shows that security techniques are designed based on video content. This kind of security methods is claimed to be far superior as they concentrate not only on watermarking but also on synchronization of watermark. In this chapter, we present a comprehensive review of multimedia security techniques emphasizing on their applicability, scope, and shortcomings especially when applied to high-definition multimedia data. Problematic issues of intelligent techniques in signal processing for multimedia security and outlook for the future research are discussed too. The major goal of the paper was to provide a comprehensive reference source for the researchers involved in designing multimedia security technique, regardless of particular application areas.

In advanced intelligent transport systems, detection of the vehicles has become very popular in the traffic area and also to identify the density of the vehicles in that particular area. As per the survey background subtraction is identified as one of the best approaches in identifying the vehicles for static camera. An improvised background subtraction model is adopted, wherein it works for real time tracking and also solves the problems of shadow detection. In background subtraction each pixel is updated with update equations. A component labeling technique is introduced after background subtraction to label the different objects so as to bifurcate between the two objects and each region is labelled with the different label values. Detections of the moving vehicles are identified and the density of vehicles travelling in the sight of the camera is determined