Defected Bitter Gourd Detection Using Convolutional Neural Network; A Computer Vision Approach to Reduce Cost and Time
- Resource Type
- Conference
- Authors
- Hasan, Md. Mehedi; Alam, Khairul; Ahmed Diganta, Md. Newaz; Nur, Arafat Ullah; Habib, Md. Tarek; Ahmed, Farruk
- Source
- 2021 12th International Conference on Computing Communication and Networking Technologies (ICCCNT) Computing Communication and Networking Technologies (ICCCNT), 2021 12th International Conference on. :1-6 Jul, 2021
- Subject
- Communication, Networking and Broadcast Technologies
Computing and Processing
Fields, Waves and Electromagnetics
Robotics and Control Systems
Signal Processing and Analysis
Deep learning
Presses
Pandemics
Computational modeling
Data models
Minerals
Magnesium
Image Classification
Deep Learning(DL)
CNN
Object Detection
Prediction
- Language
Bitter Gourd: Botanical Title is Memordica charantia L. Bitter gourd is known for its usage for medicinal purposes in Asian countries. Bitter gourd could be a wealthy source of vitamins and minerals. It contains press, magnesium, potassium, and vitamins like A and C. Different anti-oxidants and anti-inflammatory compounds are showed in Bitter gourd. Separating diseased vegetables from healthy ones of medium size vegetables is one tough job as well as costly for farmers. We propose a system which detects diseased or even defected bitter Gourd. We used Deep Learning (DL) to achieve the goal. Image processing is one of the most common yet interesting fields of DL. We used Convolutional Neural Network(CNN) to process images of Bitter Gourd. We created three different models by diversifying Convolutional Layer by number and its value. We are giving farmers a brief instrument utilizing Deep Learning that makes a difference select the non-absconded thing. The world is developing with AI within the future. So why not an agriculturist they are moreover utilizing this digitalization. We got some interesting results through the process and choose the model that performed best without test result and evaluation result over-fitting. It's M3 model with 99.70% accuracy.