College of Engineering, Pune

About: College of Engineering, Pune is a based out in . It is known for research contribution in the topics: Computer science & Sliding mode control. The organization has 4264 authors who have published 3492 publications receiving 19371 citations. The organization is also known as: COEP.

Novel Technique in Block Truncation Coding Based Feature Extraction for Content Based Image Identification

Abstract: Feature vector extraction has been the key component to define the success rate for content based image recognition. Block truncation coding is a simple technique which has facilitated various methods for effective feature vector extraction for content based image recognition. A new technique named Sorted Block Truncation Coding (SBTC) has been introduced in this work. Three different public datasets namely Wang Dataset, Oliva and Torralba (OT-Scene) Dataset and Caltech Dataset consisting of 6,221 images on the whole was considered for evaluation purpose. The technique has stimulated superior performance in image recognition when compared to classification and retrieval results with other existing techniques of feature extraction. The technique was also evaluated in lossy compression domain for the test images. Various parameters like precision, recall, misclassification rate and F1 score has been considered to evaluate the performances. Statistical evaluations have been carried out for all the comparisons by introducing paired t test to establish the significance of the findings. Classification and retrieval with proposed technique has shown a minimum of 14.4 % rise in precision results compared to the existing state-of-the art techniques.

Palm print identification using fractional coefficient of transformed edge palm images with Cosine, Haar and Kekre transform

Abstract: Paper presents performance comparison of palm print identification techniques based on fractional coefficients of transformed palm print edge image using three transforms namely Cosine, Haar and Kekre In transform domain, the energy of image gets concentrated towards low frequency region; this characteristic of image transforms is used here to reduce the feature vector size of palm print images by selecting these low frequency coefficients in transformed edge images Three image transforms applied on palm print edge image and 7 ways of taking fractional coefficients along with five different edge detection methods give total 105 variation of proposed palm print identification method Experimentation is done on a test bed of 1000 palm print images (500 left and 500 right)Genuine acceptance rate is considered for performance comparison The experimental results in Cosine and Haar transform have shown performance improvement in palm print identification using fractional coefficients of transformed images In all edge detection methods, canny is proven to be better In all Cosine transform gives best performance having maximum GAR value for canny edge detection method with 0097% of fractional coefficients considering both left and right palm print images

Synergistic effects of graphene nanoplatelets on X-band electromagnetic interference shielding, thermal expansion and thermal stability of poly(ether-ketone) based nanocomposites

Abstract: In this work, the electromagnetic interference shielding effectiveness (EMI-SE) of the poly(ether-ketone) (PEK)-graphene nanoplatelets (GNP) nanocomposites fabricated by planetary ball mill followed by hot pressing were investigated in X-band (8.2–12.4 GHz). A percolation threshold of about 0.4 vol% GNP was obtained. The electrical conductivity was increased to about 0.02 S/cm with an EMI-SE of ~33 dB for 1 mm thick 5 vol% GNP filled PEK nanocomposite. This higher value is corresponding to more than 99.95% blocking of the EMI. The EMI-SE increases with increasing thickness of the nanocomposite. The thermal stability and the char yield of the nanocomposites reinforced with 5 vol% GNP were found to increase to 570 °C and to 61.6%, respectively. The dimensional stability of the nanocomposites was also increased compared to neat PEK.

Supervised Machine Learning Algorithm for Detection of Cardiac Disorders

Abstract: The echocardiogram is the technique which is used in the diagnosis of most of the heart related diseases. Echocardiogram contains less information hence the diagnosis of the disease from the Echocardiogram videos is time consuming task. It required more human efforts to make a decision. Hence the automatic approach to detect the cardiovascular diseases with minimum computing time and high accuracy is necessary. In this approach, the automatic dilated cardiomyopathy (DCM) and Atrial Septal Defect (ASD) disease detection using machine learning approach has been proposed. The database contains the ultrasound videos of ASD, DCM and normal cases. The features extracted from the image and the classified the extracted features using supervised support vector machine algorithm. The proposed system achieved the accuracy of 98.30%.

Comparative Study of Ambient Air Quality Prediction System Using Machine Learning to Predict Air Quality in Smart City

Abstract: It is a herculean task to predict air quality of a particular area due to indefinite characteristics. As air pollution is a complex mixture of toxic air components that include ozone (O3), particulate matter 2.5_m (PM2:5), SO2, RSPM, SPM and nitrogen dioxide (NO2). These small particles penetrate deep into the alveoli as far as the bronchioles, interfering with a gas exchange within the lungs. Though research is being conducted in environmental science to evaluate the severe impact of particulate matters on public health. The capital city of Maharashtra, Nagpur is used as a case study since nearly ten thousand motor vehicles are being registered in Nagpur on a monthly basis contributing exponentially to air pollution. Various machine Learning-based algorithms are checked to compare and to find out the predictive analysis using available dataset. After comparing seven different machine learning algorithms, Boosted Random Forest algorithm was found out to be the most accurate predictive algorithm, with the maximum coefficient of determination and less mean absolute error.