Rotten Vegetable Detection

Project Title

Rotten Vegetable Detection

Project Area of Specialization Artificial IntelligenceProject Summary

We propose a computer-vision based technique for real-time and efficient classification and
quality assessment of vegetables by detecting not only their external surface characteristics but
also their internal defects with the help of hyper spectral imaging as in many cases hyper spectral
imaging is proved to be better than the usual RGB imaging. This non-destructive food quality
assessment system performing automated grading and sorting of vegetables using deep-learning
techniques will be beneficial for both agricultural and food industries.

Project Objectives

Objectives:
Following are the objectives of our proposed project:
1. To provide a real-time efficient mechanism to differentiate between high and low quality food
such as fruits and vegetables.
2. To achieve a cost-effective solution for the inspection and quality evaluation of food materials.

Project Implementation Method

Deep learning and Image Processing based techniques alongwith automatic grading and sorting of data. Images are captured by hyperspectral image capturing cameras and then image processing algorithms will be applied on them to classify on the basis of wellness of food materials and train our classifier and build its accuracy over time.

Benefits of the Project

Pakistan is an agricultural country and it's 65% of the population is dependent on agriculture.
Pakistanis major intake is agricultural products and food that impacts their health. By our project,
we could discriminate between low quality and good quality vegetables and could help them refrain
from low-quality vegetables and food made by them. We could implement this on an industrial scale
that would benefit the whole of Pakistan.

Technical Details of Final Deliverable

We are using the latest object detection and segmentation to detect and classify our vegetables.
Also using deep regression to achieve grading of the quality of vegetables.

Final Deliverable of the Project HW/SW integrated systemType of Industry IT , Food Technologies Artificial Intelligence(AI)Sustainable Development Goals Good Health and Well-Being for People, Decent Work and Economic Growth, Responsible Consumption and ProductionRequired Resources

Item Name	Type	No. of Units	Per Unit Cost (in Rs)	Total (in Rs)
			Total in (Rs)	50000
Hyperspectral imaging camera	Equipment	2	25000	50000