Multi-Camera Vision Inspection System Development (Industrial Vision Qt Application, ROI-Based Inspection)
Budget: ₹35,000 – ₹75,000 INR
Summary
We are building a desktop application (Windows/Linux) for a multi-camera vision inspection system (1-4 cam). The goal is to inspect images based on a saved reference pattern using ROI-based matching and classify each trigger as "Good" or "Bad" in under 50ms.
Please read carefully. Only apply if you have experience with industrial camera SDKs and Qt-based real-time applications.
Project Scope:
1. The application should connect to one industrial camera (GigE Vision or USB3) and display the live feed in a Qt-based UI.
2. The user should be able to draw one rectangular ROI on the live feed to capture and save a reference image.
3. The software should automatically generate a slightly larger "search area" around the ROI to allow for minor object shifts. If the object appears outside this area, it should be marked as "Bad".
4. On each hardware trigger, the app should:
- Capture a new image
- Compare only the ROI/search area against the saved reference
- Decide Good/Bad using a configurable tolerance percentage
- Complete the decision process within 50 milliseconds
5. The app must save the reference ROI and image even after shutdown. It should automatically load them on restart unless a new ROI is defined.
6. All inspection results should be logged with timestamps and Pass/Fail status. A CSV export or on-screen summary of results should be available.
Proof of Concept (POC) Requirement:
Before assigning the full project, we require a working demo (POC):
- A Qt desktop application that connects to an industrial camera (GigE or USB3)
- Shows the live camera feed in a UI
- Includes basic processing (e.g. grayscale conversion)
- Candidates who complete this POC will be shortlisted for the full MVP development.
To Apply:
Please include the following:
1. A brief summary of a similar machine vision or camera integration project you’ve completed
2. Any GitHub repos, screenshots, or videos (if project is private, describe your contribution)
Your availability and estimated cost for the MVP
Bonus if you have:
1. Experience with GenICam/GenTL-based SDKs
2. Worked on multi-camera inspection or synchronized image acquisition
3. Looking forward to collaborating with the right candidate.
Tech Stack: Qt (C++), OpenCV, Industrial Camera SDK (GigE or USB3) any other image processing libraries
We are building a desktop application (Windows/Linux) for a multi-camera vision inspection system (1-4 cam). The goal is to inspect images based on a saved reference pattern using ROI-based matching and classify each trigger as "Good" or "Bad" in under 50ms.
Please read carefully. Only apply if you have experience with industrial camera SDKs and Qt-based real-time applications.
Project Scope:
1. The application should connect to one industrial camera (GigE Vision or USB3) and display the live feed in a Qt-based UI.
2. The user should be able to draw one rectangular ROI on the live feed to capture and save a reference image.
3. The software should automatically generate a slightly larger "search area" around the ROI to allow for minor object shifts. If the object appears outside this area, it should be marked as "Bad".
4. On each hardware trigger, the app should:
- Capture a new image
- Compare only the ROI/search area against the saved reference
- Decide Good/Bad using a configurable tolerance percentage
- Complete the decision process within 50 milliseconds
5. The app must save the reference ROI and image even after shutdown. It should automatically load them on restart unless a new ROI is defined.
6. All inspection results should be logged with timestamps and Pass/Fail status. A CSV export or on-screen summary of results should be available.
Proof of Concept (POC) Requirement:
Before assigning the full project, we require a working demo (POC):
- A Qt desktop application that connects to an industrial camera (GigE or USB3)
- Shows the live camera feed in a UI
- Includes basic processing (e.g. grayscale conversion)
- Candidates who complete this POC will be shortlisted for the full MVP development.
To Apply:
Please include the following:
1. A brief summary of a similar machine vision or camera integration project you’ve completed
2. Any GitHub repos, screenshots, or videos (if project is private, describe your contribution)
Your availability and estimated cost for the MVP
Bonus if you have:
1. Experience with GenICam/GenTL-based SDKs
2. Worked on multi-camera inspection or synchronized image acquisition
3. Looking forward to collaborating with the right candidate.
Tech Stack: Qt (C++), OpenCV, Industrial Camera SDK (GigE or USB3) any other image processing libraries
Related categories:
C Programming
Windows Desktop
Linux
Software Architecture
MySQL
C++ Programming
Qt
OpenCV