About project
An artificial intelligence platform provides surveyors with automatic quality control and categorization for fruits and vegetables.
Facts
6 team members including 3 software developers
80k USD Budget
Revolutionary technology applied
Customer
-
Our customer is a former top manager of one of the largest Russian retailers, leads a world-class surveyor and logistics professionals. Spellsystems delivered a solution with the following specification:
Goal

- Accuracy of 99%
- Defects Detection
- Product Quality Detection
- Quality Category Detection
- Databases conducted with the accordance with the UNECE and OECD standards
Challenge
- The software automates the entire process of dataset and feature management, ANN training pipeline, and deployment.
- Neural network algorithms assess the quality of fresh fruits and vegetables.
- The software covers all the essential surveyors' business processes. including analysis of products, shipment, and statistics.
- Databases of examination results are saturated. All the examinations meet the United Nations Economic Commission for Europe (UNECE) and the Organization for Economic Cooperation and Development (OECD) standards.
Results
1
A scanner for obtaining high-quality photographs has been developed
2
A unit of trade professionals has been formed to assess the quality of fruits and vegetables
3
Data collection for machine learning is ongoing
4
Developed software to assess the quality of fruits and vegetables
5
The automated process of forming reports has been developed
6
The virtual personal Cabinet of the expert-receiver has been developed
Presentation
Technologies
Basic technologies
1
C#, Microsoft.NET
Python
Python
Artifical intelligence
2
Convolutional neural networks, ResNET
RetinaNET
RetinaNET
Platforms
Web applications
Desktop applications
Mobile applications
Embeded
Server applications