Browsing by Author "Mahfuzullah, Md."

Filter results by typing the first few letters
Now showing 1 - 2 of 2
  • No Thumbnail Available
    Item
    A Computer Vision Based Food Recognition Approach for Controlling Inflammation to Enhance Quality of Life of Psoriasis Patients
    (2021 International Conference on Information Technology (ICIT), IEEE, 2021-07-26) Hridoy, Rashidul Hasan; Akter, Fatema; Mahfuzullah, Md.; Ferdowsy, Faria
    Deep learning becomes the spotlight in computer vision based recognition approaches in recent years. Psoriasis affects people of all ages around the world and causes inflammation on the skin with significant systemic disability and illness. Inflammatory foods increase inflammation rapidly, patients can easily control inflammation to enhance the quality of life by eliminating these foods from their everyday diet. This paper addresses a rapid food recognition approach to assist psoriasis patients to recognize fifteen highly inflammatory foods. Using image augmentation techniques, a dataset of 41250 images of different inflammatory foods have generated from 10000 images. AlexNet, VGG16, and EfficientNet-B0 have used in this study using the transfer learning approach, and EfficientNet-B0 has achieved the highest accuracy of 98.63% under the test set of 5250 images. AlexNet and VGG16 have achieved 87.22% and 93.79% accuracy, respectively. EfficientNet-B0 has consumed the lowest time in recognizing unseen images compared to others.
  • Thumbnail Image
    Item
    Cow Disease Recognition Using Convolutional Neural Network
    (Daffodil International University, 2022-01-04) Mahfuzullah, Md.; Bin Hasan, Asif
    In the South Asian region, cow illness is a regular occurrence. Every season, poor rural residents, as well as dairy farm owners, face the same issue. Dairy farming is a large and developing agricultural sector in Bangladesh - a country of South Asia. A large portion of the population relies on cattle for a living. Cow milk is a source of pure nutrition in Bangladesh hence cows provide the bulk of milk and meat. Every year, farmers lose a lot of money due to cow diseases. Cow diseases reduce the amount of milk and meat produced. Recognizing cow diseases has thus become a critical undertaking to be fulfilled. In recent years, deep learning has been gaining a lot of popularity because of its superior accuracy when trained with a lot of data. Using deep learning, we will develop a system to recognize cow diseases. We have begun by reading many online publications, journals, and relevant studies before heading out to the field to tour dairy farms and village habitats. Brucellosis, lumpy skin disease, foot & mouth disease, and mastitis diseases are the four most frequently occurring cow diseases in Bangladesh. Then we have collected pictures of both healthy and diseased cows. We have proceeded to our dataset with five classes after gathering all of the picture data. To recognize cow diseases, we have used three deep learning pre-trained models with our dataset. We have achieved satisfactory results with the MobileNetV2 by attaining accuracy of 95.43%. For this study, we have also used two more pre-trained models, Alexnet and VGG16 which exhibit the accuracy of 86.27% and 90.85% respectively. The MobileNetV2 has not only performed the best in terms of accuracy but also some other indicative performance metrics like sensitivity, specificity, and precision. This research can help us recognize cow diseases quickly and avoid the unexpected loss of our livestock. As well, it also has a bright future in both the domestic and international markets.