Plant disease diagnosis using deep transfer learning architectures- VGG19, MobileNetV2 and Inception-V3
This thesis is submitted in partial fulfillment of the requirements for the degree of Bachelor of Science in Computer Science and Engineering, 2022.
| Main Authors: | , , |
|---|---|
| 其他作者: | |
| 格式: | Thesis |
| 语言: | English |
| 出版: |
Brac University
2022
|
| 主题: | |
| 在线阅读: | http://hdl.handle.net/10361/17334 |
| id |
10361-17334 |
|---|---|
| record_format |
dspace |
| spelling |
10361-173342022-09-27T21:02:17Z Plant disease diagnosis using deep transfer learning architectures- VGG19, MobileNetV2 and Inception-V3 Kobra, Khadija-Tul Suham, Rahmatul Rashid Fairooz, Maisha Uddin, Jia Rodosh, Ahanaf Hassan Department of Computer Science and Engineering, Brac University Plant diseases VGG19 MobileNetV2 Inception-v3 Crop Cognitive learning theory (Deep learning) Plant diseases--Diagnosis This thesis is submitted in partial fulfillment of the requirements for the degree of Bachelor of Science in Computer Science and Engineering, 2022. Cataloged from PDF version of thesis. Includes bibliographical references (pages 28-29). The importance of a tree’s involvement in human life and the environment cannot be overstated. Plants, like humans and animals, are susceptible to disease. Many vegetation diseases might impair a plant’s healthy development. For accurate identification and treatment of plant pathogens, precise detection of those chronic conditions is essential. This paper represents 3 deep learning approaches to distinguish and classify plant diseases by analyzing the leaf of a given plant. We worked on late leaf curl, leaf spot, mosaic virus, black rot, powdery mildew, common rust, bacterial spot, leaf scorch, syndromes of late and early blight of crops similar to corn, potato, tomato, squash, pepper, cherry, grape, orange, strawberry, apple etc. The proposed strategy improves disease identification and classification of deformed collected leaves. The model performs its function by categorizing images into two groups, diseased and healthy. Moreover, deep learning architectures are made up of several processing layers that learn the data visualizations with discrete levels of abstraction. Collecting data sets is one of the most crucial steps to creating any recognition system. Labeling an image means pinpointing the subject we will be trying to find, Training the algorithms through those images to detect the subjects is critical in detecting diseases. In this research paper, to detect diseases from images, firstly, we collected data sets containing more than 87,123 images of plant diseases. After that, we labeled those images in 38 labels and we used VGG19 model on those images to train the model to detect diseases from the images given to it, afterwards two deep learning models MobileNetV2 and Inception-v3 was used to detect diseases which provided us with 94.21%, 97.93% and 98.52% accuracy respectively. In short, we’re using three deep learning models and comparing the accuracy rate on a huge data set with 38 classes which will help the masses to detect abnormalities in plants. It will also help the harvesters related to the agricultural works find the contagion in their cultivated crops further to develop our horticulture sector and our farmers’ situation. Khadija-Tul-Kobra Rahmatul Rashid Suham Maisha Fairooz B. Computer Science and Engineering 2022-09-27T05:12:47Z 2022-09-27T05:12:47Z 2022 2022-05 Thesis ID 17301018 ID 18101681 ID 18101067 http://hdl.handle.net/10361/17334 en Brac University theses are protected by copyright. They may be viewed from this source for any purpose, but reproduction or distribution in any format is prohibited without written permission. 29 pages application/pdf Brac University |
| institution |
Brac University |
| collection |
Institutional Repository |
| language |
English |
| topic |
Plant diseases VGG19 MobileNetV2 Inception-v3 Crop Cognitive learning theory (Deep learning) Plant diseases--Diagnosis |
| spellingShingle |
Plant diseases VGG19 MobileNetV2 Inception-v3 Crop Cognitive learning theory (Deep learning) Plant diseases--Diagnosis Kobra, Khadija-Tul Suham, Rahmatul Rashid Fairooz, Maisha Plant disease diagnosis using deep transfer learning architectures- VGG19, MobileNetV2 and Inception-V3 |
| description |
This thesis is submitted in partial fulfillment of the requirements for the degree of Bachelor of Science in Computer Science and Engineering, 2022. |
| author2 |
Uddin, Jia |
| author_facet |
Uddin, Jia Kobra, Khadija-Tul Suham, Rahmatul Rashid Fairooz, Maisha |
| format |
Thesis |
| author |
Kobra, Khadija-Tul Suham, Rahmatul Rashid Fairooz, Maisha |
| author_sort |
Kobra, Khadija-Tul |
| title |
Plant disease diagnosis using deep transfer learning architectures- VGG19, MobileNetV2 and Inception-V3 |
| title_short |
Plant disease diagnosis using deep transfer learning architectures- VGG19, MobileNetV2 and Inception-V3 |
| title_full |
Plant disease diagnosis using deep transfer learning architectures- VGG19, MobileNetV2 and Inception-V3 |
| title_fullStr |
Plant disease diagnosis using deep transfer learning architectures- VGG19, MobileNetV2 and Inception-V3 |
| title_full_unstemmed |
Plant disease diagnosis using deep transfer learning architectures- VGG19, MobileNetV2 and Inception-V3 |
| title_sort |
plant disease diagnosis using deep transfer learning architectures- vgg19, mobilenetv2 and inception-v3 |
| publisher |
Brac University |
| publishDate |
2022 |
| url |
http://hdl.handle.net/10361/17334 |
| work_keys_str_mv |
AT kobrakhadijatul plantdiseasediagnosisusingdeeptransferlearningarchitecturesvgg19mobilenetv2andinceptionv3 AT suhamrahmatulrashid plantdiseasediagnosisusingdeeptransferlearningarchitecturesvgg19mobilenetv2andinceptionv3 AT fairoozmaisha plantdiseasediagnosisusingdeeptransferlearningarchitecturesvgg19mobilenetv2andinceptionv3 |
| _version_ |
1814308759161274368 |