Chittagong University of Engineering & Technology

About: Chittagong University of Engineering & Technology is a education organization based out in Chittagong, Bangladesh. It is known for research contribution in the topics: Computer science & Renewable energy. The organization has 1200 authors who have published 1444 publications receiving 10418 citations. The organization is also known as: Engineering College, Chittagong & Bangladesh Institute of Technology, Chittagong.

Vulnerability assessment and construction recommendations of local houses in the cyclone prone coastal areas of Bangladesh

Abstract: Bangladesh has approximately 710 km long coastline at its southern part. This huge coastal region is very much prone to tidal surge and cyclones; and approximately in every 5 years devastating cyclonic storm hits these coastal areas. The entire coastline suffers serious damage of locally built traditional or non-engineered houses in almost every year. Therefore, this study investigates a proper wind-resistant housing concept, so that people can build houses which are structurally strong to resist the high speed winds and as well be within their affordable financial limits. In order to carry out this study, existing housing data in the extreme wind-swept coastal areas of Chittagong, Cox-bazar and Patuakhali districts of Bangladesh was collected and analyzed to categorize different housing patterns. Some houses were found very poor in condition from the construction point of view and were not recommended for further strengthening. Some houses were found good in condition but had been constructed without following any engineering practices. These types of houses were considered for further strengthening to resist high speed winds that are prevalent in the coastal areas of Bangladesh. A specific pattern of house was analyzed using finite element software ANSYS 11.0. Thereafter, different types of strengthening techniques were applied and analyzed them for obtaining stress and deflection data. Results show that the deflections reduced by 95% when simple strengthening techniques (adding tie and bracing at some important points) were applied to the existing houses. This article also provides some guidelines for construction of houses in cyclone-prone areas.

Comparative Study Between Decision Tree, SVM and KNN to Predict Anaemic Condition

Abstract: Anemia, a disease which is caused by an inadequacy of hemoglobin or red blood cells in the blood. It is very risky at the time of pregnancy, menstruation and in ICU sometimes causing death. So, it is a need of hemoglobin and detects anemia quickly. Usually, doctors examine the eye conjunctiva color and confirmed by a blood test which is painful, time-consuming and costly. In this study, a total of 104 people (54 males and 50 females) are collected with their clinical blood hemoglobin level, anemic condition and taken palpebral conjunctiva image. The images are captured with a cell phone camera of good resolution. By using the images, the percentage of the red, green and blue pixels are extracted in MATLAB, image processing method. Taking those features, the Hemoglobin level is plotted. A total of 81 data is taken for training purposes and 23 data for testing. For Anemia detection, the 81 data are trained with a used different classifier such as Linear SVM, Coarse Tree, and Cosine KNN and have been got highest accuracy of 82.61% in Decision Tree (Coarse) by testing 23 data.

CricShotClassify: An Approach to Classifying Batting Shots from Cricket Videos Using a Convolutional Neural Network and Gated Recurrent Unit

Abstract: Recognizing the sport of cricket on the basis of different batting shots can be a significant part of context-based advertisement to users watching cricket, generating sensor-based commentary systems and coaching assistants. Due to the similarity between different batting shots, manual feature extraction from video frames is tedious. This paper proposes a hybrid deep-neural-network architecture for classifying 10 different cricket batting shots from offline videos. We composed a novel dataset, CricShot10, comprising uneven lengths of batting shots and unpredictable illumination conditions. Impelled by the enormous success of deep-learning models, we utilized a convolutional neural network (CNN) for automatic feature extraction, and a gated recurrent unit (GRU) to deal with long temporal dependency. Initially, conventional CNN and dilated CNN-based architectures were developed. Following that, different transfer-learning models were investigated—namely, VGG16, InceptionV3, Xception, and DenseNet169—which freeze all the layers. Experiment results demonstrated that the VGG16–GRU model outperformed the other models by attaining 86% accuracy. We further explored VGG16 and two models were developed, one by freezing all but the final 4 VGG16 layers, and another by freezing all but the final 8 VGG16 layers. On our CricShot10 dataset, these two models were 93% accurate. These results verify the effectiveness of our proposed architecture compared with other methods in terms of accuracy.