Computer Vision System

Advanced methods for real-time object recognition based on deep neural networks technology

Detection of objects in an image is an actual problem in computer vision. Given the current trend towards automation, fast and efficient systems for object recognition are required.
And thanks to the growth of hardware performance and the emergence of large data sets with various images, computer vision algorithms effectively solve the problems of image classification, object detection and image segmentation.

The program is designed to recognize and classify visual images in images and video sequences using a convolutional neural network, which increases the speed and quality of detection.
As a data source, you can select "Camera" - recording from a camera in real time, "Video" - uploading a video file, "Pictures" - uploading multiple images (you must select a folder that will contain several images), "Picture" - uploading one image.
Supported video formats: mkv, flv, avi, wmv, mp4, mpeg, flv. Supported image formats: jpg, png, jpeg, bmp.
It is possible to select the type of computing resources of the computer for data processing: CPU or GPU.
And you can also choose the type of neural network for data processing: YOLO and YOLOv2-tiny, where YOLOv2-tiny is a faster version.
One of the modern detectors is the YOLO (You Only Look Once) convolutional neural network. Thanks to its architecture, it only needs to “look” at the image once, which significantly reduces the time required to detect an object in the frame


The result of the software operation is the definition of objects in the picture/frame or video stream. That is, the algorithm or neural network determines the object and records its position and (parameters of the rectangles around the objects). It also links information from previous frames in such a way as not to lose the object, or make it unique.

The program provides an opportunity of:

1. Automatic medical image processing
2. Monitoring and predicting the patient's condition
3. Auto-recognition of moving objects
4. Implementation of control over the production and quality of manufactured products
5. Object identification
6. Pattern recognition
7. Analysis of conducted surveys
8. Predictive ratings
9. Face identification
10. License plate identification
11. Personal identification
12. Text recognition
13. Image recognition
14. Access control of employees
15. Food accounting
16. Recognition of objects of goods on the shelves for subsequent optimization of the location of goods
17. For detecting fires from drones
18. Gas leaks and more
Contact Us
We're ready to help
Please note that by clicking the "Submit" button you agree to our Privacy Policy