Browsing by Author "Ashfia Binte Habib"

Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    Open Access
    COVID-19 Diagnosis and Classification from CXR Images based on Vision Transformer (ViT)
    (North South University, 2021) Mahbubur Rahman; Shihabur Rahman Samrat; Abdullah Al Ahad; Ashfia Binte Habib; 1731134042; 1731574042; 1731496042
    The COVID-19 pandemic is far from over, and the current primary method of diagnosis is Reverse Transcription Polymerase Chain Reaction (RT-PCR). Although RT-PCR is reliable, it is known to have a long turnaround time and high false-negative rates that can severely hinder the accuracy of diagnosis. Alongside RT-PCR, Rapid Antigen Tests (RAT) are also used, but they have much lower accuracy than RT-PCR. Motivated by the flaws of the current diagnosis methods, we present a Vision Transformer-based classifier for the successful diagnosis and classification of COVID-19 using chest X-Ray (CXR) images. A 15000-sample CXR dataset was compiled, which consisted of 5000 CXRs per class. Afterwards, a Vision Transfer (ViT) was fine-tuned on the dataset. ResNet-50 and DenseNet121 were used as baseline models. It is observed that the Vision Transformer-based model had the highest classification accuracy of 96.2% with an F1 score of 0.965 and the average precision and recall of 0.9617 and 0.962, respectively. This study demonstrates the adequacy of the ViT for the identification and classification of COVID-19 and Pneumonia.
  • Thumbnail Image
    Item
    Open Access
    Krishi House An IoT based smart greenhouse
    (North South University, 2021) Moumita Ferdoushi; Shorony Mehtaz Ahmed; Ashfia Binte Habib; 1631125045; 1331066045
    In Bangladesh, one of the main economic activities lies in the agricultural sector. Although the good presence of resources does not produce results equal to the availability. The cause of scarcity and incompetent use of technology, deficiency of knowledge and awareness among the agrarians, methods which are time-consuming, hard labor, and more expensive in the long run. A Greenhouse is a technical approach in which farmers in rural areas will be benefitted by automatic monitoring and control of the greenhouse environment. The project focuses on the use of IoT in the greenhouse for environment monitoring and control by implementing low-cost, space, and effort strategies. A system that is efficient in terms of power and water usage while keeping the system architecture and design modest. Krishi House is semi-automated making human supervision redundant. With the help of a temperature sensor, soil moisture, pH and humidity sensor, light sensor the system will take inputs from the sensors to the microcontroller and microprocessor of the greenhouse where the algorithm will give desired output in form of irrigation, growth light, and system alert along with sending data wirelessly to the user. Power efficiency is attained by the use of solar energy when acquired; cost and space efficiency is achieved with a meticulous selection of components, materials, and horticulture technique. The goal is to have a competent semiautomated eco-system that supports different types of vegetation which will provide a better yield than the traditional farming method without the hard labor keeping cost and power use at least. The data collected throughout the experiment by the microcontroller and the microprocessor are attained wirelessly in a cloud database and can also be seen by a mobile app at the user’s convenience. The goal is to achieve a High-tech greenhouse with automation without the extra cost. The use of IoT and technological advancement will improve the agricultural prospect drastically in Bangladesh which shall be economically accommodating.