Abstract
This thesis is divided into two parts:Part I: Analysis of Fruits, Vegetables, Cheese and Fish based on Image Processing using Computer Vision and Deep Learning: A Review. It consists of a comprehensive review of image processing, computer vision and deep learning techniques applied to carry out analysis of fruits, vegetables, cheese and fish.This part also serves as a literature review for Part II.Part II: GuavaNet: A deep neural network architecture for automatic sensory evaluation to predict degree of acceptability for Guava by a consumer. This part introduces to an end-to-end deep neural network architecture that can predict the degree of acceptability by the consumer for a guava based on sensory evaluation.
View Research Publication:
Overview
As a part of research during my Master of Science at Rochester Institute of Technology I Developed a novel deep neural network architecture to automate sensory evaluation of Guava, creating a non-destructive quality assessment method that could be extended to other fruits. This research aimed to replace traditional destructive testing methods with a computer vision-based approach, improving efficiency and reducing waste in the agricultural supply chain.
Methodology
- Explored various pre-trained Convolutional Neural Network (CNN) architectures such as VGG-16, ResNet-18, ResNet-50, YOLOv4.
- Pipeline consisted of YOLOv4 for object detection and localization and CNN + Custom Fully-Connected Network (FCN) for grading and sensory evaluation.
Architecture
Output
Technologies
Python, OpenCV, PyTorch, YOLO V4, ResNet-18, ResNet-50, VGG-16