Object recognition and classification

The ORC algorithm uses neural networks to recognise and detect objects in the camera image. It is possible to recognise static objects by analysing the entire image or moving objects in the image from a static camera. Objects in real time can also be recognised or camera recordings post-processed and analysed. Depending on the assignment specification, algorithms can be taught to recognise and classify new objects that the neural network has not yet recognised.

It is also possible to specify various classification criteria, such as the number of frames in which the object must be located or the recognition success threshold. The ORC algorithm can be used to recognise moving objects in security systems, but also to classify obstacles in front of a robot. The output is the name of the object marked by the bounding box, the recognition success and the position in the image.

Input data

Photos, video, dataset

Types of objects

Static, dynamic

Output data

Object name, recognition success, position in the image

Output format

XML, JSON and others


People detection

A device equipped with the ORC algorithm can detect and distinguish a person from a photo or video and for example alert to it.

Face recognition

Using the ORC algorithm, it is also possible to recognise faces that the system knows, their gender, age or even their mood.

Object recognition

The ORC algorithm can also distinguish objects such as cars, bicycles, animals, common and non-traditional objects such as weapons.

Visual recognition

With the ORC algorithm it is possible to detect a car registration number, distinguish product packaging or extract text from an image.

Environment recognition

The ORC algorithm can also be used to navigate a device that uses it to segment a path or detect walls and navigate between them.

Custom objects

Neural networks in the ORC algorithm can be trained for various objects. Detection success depends on the dataset size and quality.


The ORC algorithm consists of several software solutions, its practical deployment requires hardware depending on the purpose of the system.

RoboTech Vision logo

The company focuses on the development of autonomous robots with AI elements. It strives to develop universal solutions for various tasks, industries and environments.


AON algorithm

AVN algorithm

ADN algorithm

ORC algorithm




RTV sensor Box


RoboTech Vision Ltd.
Červený kameň 61
900 89 Častá

Pin It on Pinterest

Share This