{"title":"A Novel Approach of Image Fusion Techniques using Ant Colony Optimization","authors":"J. Kulkarni, R. Bichkar","doi":"10.35940/ijitee.h9241.0610821","DOIUrl":"https://doi.org/10.35940/ijitee.h9241.0610821","url":null,"abstract":"Ant Colony Optimization (ACO) is a relatively high approach for finding a relatively strong solution to the problem of optimization. The ACO based image fusion technique is proposed. The objective function and distance matrix is designed for image fusion. ACO is used to fuse input images at the feature-level by learning the fusion parameters. It is used to select the fusion parameters according to the user-defined cost functions. This algorithm transforms the results into the initial pheromone distribution and seeks the optimal solution by using the features. As to relevant parameters for the ACO, three parameters (α, β, ρ ) have the greatest impact on convergence. If the values of α, β are appropriately increased, convergence can speed up. But if the gap between these two is too large, the precision of convergence will be negatively affected. Since the ACO is a random search algorithm, its computation speed is relatively slow.","PeriodicalId":23601,"journal":{"name":"VOLUME-8 ISSUE-10, AUGUST 2019, REGULAR ISSUE","volume":"109 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2021-06-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83517024","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Vehicular Security: Drowsy Driver Detection System","authors":"Pranavi Pendyala, Aviva Munshi, Anoushka Mehra","doi":"10.35940/ijeat.e2751.0610521","DOIUrl":"https://doi.org/10.35940/ijeat.e2751.0610521","url":null,"abstract":"Detecting the driver's drowsiness in a consistent\u0000and confident manner is a difficult job because it necessitates\u0000careful observation of facial behaviour such as eye-closure,\u0000blinking, and yawning. It's much more difficult to deal with when\u0000they're wearing sunglasses or a scarf, as seen in the data\u0000collection for this competition. A drowsy person makes a variety\u0000of facial gestures, such as quick and repetitive blinking, shaking\u0000their heads, and yawning often. Drivers' drowsiness levels are\u0000commonly determined by assessing their abnormal behaviours\u0000using computerised, nonintrusive behavioural approaches. Using\u0000computer vision techniques to track a driver's sleepiness in a\u0000non-invasive manner. The aim of this paper is to calculate the\u0000current behaviour of the driver's eyes, which is visualised by the\u0000camera, so that we can check the driver's drowsiness. We present a\u0000drowsiness detection framework that uses Python, OpenCV, and\u0000Keras to notify the driver when he feels sleepy. We will use\u0000OpenCV to gather images from a webcam and feed them into a\u0000Deep Learning model that will classify whether the person's eyes\u0000are \"Open\" or \"Closed\" in this article.","PeriodicalId":23601,"journal":{"name":"VOLUME-8 ISSUE-10, AUGUST 2019, REGULAR ISSUE","volume":"34 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2021-06-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"89956751","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"A Proposal for Sleep Scoring Analysis Designed by Computer Assisted using Physiological Signals","authors":"Hemu Farooq, Anuj Jain, V. K. Sharma","doi":"10.35940/ijeat.e2609.0610521","DOIUrl":"https://doi.org/10.35940/ijeat.e2609.0610521","url":null,"abstract":"Sleep is utterly regarded as compulsory component\u0000for a person’s prosperity and is an exceedingly important element\u0000for wellbeing of a healthy person. It is a condition in which an\u0000individual is physically and mentally at rest. The conception of\u0000sleep is considered extremely peculiar and is a topic of discussion\u0000and researchers all over the world has been attracted by this\u0000concept. Sleep analysis and its stages is analyzed to be useful in\u0000sleep research and sleep medicine area. By properly analyzing\u0000the sleep scoring system and its different stages has proven\u0000helpful for diagnosing sleep disorders. As it’s seen, sleep stage\u0000classification by manual process is a hectic procedure as it takes\u0000sufficient time for sleep experts to perform data analysis. Besides,\u0000mistakes and irregularities in between classification of same data\u0000can be recurrent. Therefore, the use of automatic scoring system\u0000in order to support reliable classification is highly in greater use.\u0000The scheduled work provides an insight to use the automatic\u0000scheme which is based on real time EMG signals and Artificial\u0000neural network. EMG is an electro neurological diagnostic tool\u0000which evaluates and records the electrical activity generated by\u0000muscle cells. The sleep scoring analysis can be applied by\u0000recording Electroencephalogram (EEG), Electromyogram\u0000(EMG), and Electrooculogram (EOG) based on epoch and this\u0000method is termed as PSG test or polysomnography test. The\u0000epoch measured has length segments for a period of 30 seconds.\u0000The standard database of EMG records was gathered from\u0000various hospitals in sleep laboratory which gives the different\u0000stages of sleep. These are Waking, Non-REM1 (stage-1), NonREM2 (stage-2), Non-REM3 (stage-3), REM. The collection of\u0000data was done for the period of 30 second known as epoch, for\u0000seven hours. The dataset obtained from the biological signal was\u0000managed so that necessary data is to be extracted from\u0000degenerated signal utilized for the purpose of study. As a matter\u0000of fact, it is known electrical signals are distributed throughout\u0000the body and is needed to be removed. These unwanted signals\u0000are termed as artifacts and they are removed with the help of\u0000filters. In this proposed work, the signal is filtered by making use\u0000of low-pass filter called Butterworth. The withdrawn\u0000characteristics were instructed and categorized by utilizing\u0000Artificial Neural Network (ANN). ANN, on the other hand is\u0000highly complicated network and utilizing same in the field of\u0000biomedical when contracted with electrical signals, acquired\u0000from human body is itself a novel. The precision obtained by the\u0000help of the procedure was discovered to be satisfactory and hence\u0000the process is very useful in clinics of sleep, especially helpful for\u0000neuro-scientists for discovering the disturbance in sleep.","PeriodicalId":23601,"journal":{"name":"VOLUME-8 ISSUE-10, AUGUST 2019, REGULAR ISSUE","volume":"49 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2021-06-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74904029","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Anomaly Detection Algorithms in Financial Data","authors":"Abhisu Jain, Mayank Arora, Anoushka Mehra, Aviva Munshi","doi":"10.35940/ijeat.e2598.0610521","DOIUrl":"https://doi.org/10.35940/ijeat.e2598.0610521","url":null,"abstract":"The main aim of this project is to understand and apply\u0000the separate approach to classify fraudulent transactions in a\u0000database using the Isolation forest algorithm and LOF algorithm\u0000instead of the generic Random Forest approach. The model will be\u0000able to identify transactions with greater accuracy and we will\u0000work towards a more optimal solution by comparing both\u0000approaches. The problem of detecting credit card fraud involves\u0000modelling past credit card purchases with the perception of those\u0000that turned out to be fraud. Then, this model is used to determine\u0000whether or not a new transaction is fraudulent. The objective of\u0000the project here is to identify 100% of the fraudulent transactions\u0000while mitigating the incorrect classifications offraud.","PeriodicalId":23601,"journal":{"name":"VOLUME-8 ISSUE-10, AUGUST 2019, REGULAR ISSUE","volume":"1 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2021-06-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75856285","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Microstrip Feed Trapezoidal Shape Antenna Array with Defected Ground Structure for S\u0000Band Applications","authors":"S. Santhanam, Thiruvalar Selvan Palavesam","doi":"10.35940/ijeat.d2497.0610521","DOIUrl":"https://doi.org/10.35940/ijeat.d2497.0610521","url":null,"abstract":"In this proposal new trapezoidal patch microstrip feed\u0000antenna array with ground defected by square shape is designed\u0000for detailed antenna parameter study in terms of return loss,\u0000VSWR, gain and radiation pattern for S band applications from 2\u0000to 3 GHz. The bandwidth and radiation properties of four\u0000radiating element arranged in 2 x 2 array has been improved by\u0000defecting half of the ground by etching square shape opposite to\u0000the vertical feed point. 30 x 70 x 1.6 mm dimension structure has\u0000been fabricated in FR4 substrate for low cost applications and\u0000performance analyzed in three different planes. With comparison\u0000of four element array with full ground, the proposed array with\u0000defected ground has proved the improvement in behavior with\u0000return loss of -34.687 dB and ideally fit with VSWR of 1.038.\u0000Parametric study with feed length and substrate thickness has also\u0000been performed optimized decision of structure dimension. This\u0000study reveals that by reducing the substrate thickness and\u0000increasing the feed length, we can improve the performance of\u0000loss reduction. The front view has been simulated with full ground\u0000and defected ground for comparison and the compared results\u0000shows that the loss reduction of -22 dB has been achieved with\u0000VSWR value of 1.03 from 2.28 for defected ground structure. The\u0000designed structure has been simulated with CST software and the\u0000comparison of simulated results has conform that the proposed\u0000structure can be used for S band application like airport\u0000surveillance radars with wide bandwidth of 120 MHz and gain of\u00003.52 dBi. Comparison has been made between the proposed\u0000antenna array and the antennas available in literature with\u0000respect to bandwidth gain, reflection coefficient and defection type\u0000for better understanding.","PeriodicalId":23601,"journal":{"name":"VOLUME-8 ISSUE-10, AUGUST 2019, REGULAR ISSUE","volume":"42 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2021-06-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"82452953","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}