endorsedOngoing

Computer Vision Implementation for Def C's UAVs

Services Covered:

Custom Software Development

Industry

Aviation

Duration

88 months

Budget

10K - 49K

Start Date

1 April 2019

Team Size

1-5 Employees

Engagement Model

Nearshore

Project Capability Score (PCS)

73/100

Good

Project Capability Score (PCS) estimates how capable Softengi is of successfully delivering a project like this, based on its past experience, track record, and ability to handle similar work. Learn more about PCS

Client

Def C

Ukraine

Project Summary

Softengi implemented computer vision software for the client's UAVs, creating onboard systems for object recognition, tracking, and monitoring using Python and Open CV. The integration improved object-recognition accuracy and enabled offline video processing; the client now uses CV-enhanced drones for utility leak monitoring and reports steady improvements in recognition accuracy.

Key Challenges

Increase the accuracy of target recognition, monitoring, and detection.
Minimize human-made errors in target monitoring.
Make drones useful across multiple purposes (logistics escorting, utility leak monitoring, agriculture, military).
Increase the amount of accurate actionable data generated daily and improve reaction speed to that data.

Project Deliverables

Computer vision software for UAV onboard plates
Object recognition, tracking, and monitoring system
Offline video processing capability (works without base signal)
Secure on-premises data storage
Web visualization interface for processed data

Project Solution

After vendor selection interviews, Softengi implemented a step-by-step computer vision plan: they improved UAV onboard plates and developed a CV software stack for object recognition, tracking, and monitoring using Python and the OpenCV library. The system supports video processing without a base signal and stores data on the client's premises to enhance data security; a web visualization interface was provided for processed visual data.

Project Outcome

Drones are actively used for leak monitoring in the utility sector.
Increased accuracy of object recognition, with steady improvement.
Effective workflow tracked via Jira and Slack and the team has met deadlines so far.

Platforms

AR/VR/XR
Web

Tech Stack

Python

Client Endorsement

Overall Review Rating

Timeliness

Cost Rating

Willing to Refer

Quality of Deliverables

“Today, the client enjoys drones enhanced with computer vision allowing for expansive monitoring and security. Softengi has also increased the accuracy of the drones' object recognition. In the ongoing project, the agency continuously indicates a deep investment in the technology of their work.”
Igor Kramarenko
CTO
Date: 08/13/2019