MRL Projects

Reading the marking on the steel ingots

This is a new project that is just starting (summer 2009). It should be done for a partner from industry. The goal is to read, by making use of a camera, the markings that are imprinted on the hot steel ingots. The result of reading will be used for production control. The conditions under which the images should be captured and the marking should be read are quite difficult. At this moment, we are at the very beginning and we are thinking about how to solve it. The project is funded by the Ministry of Industry and Trade of the Czech republic. You can learn more here (will be available soon).

Computing the car lamps light flux

In this project, we create a software for computing the light flux from the car lamps (especially headlamps). This project is being solved for one big car-parts manufacturer. Computing the light flux accurately and effectively is an important task when designing the head lamps. The accurate results are necessary especially in case that new light emitters and new materials are to be used. Since the computation is time expensive on usual personal computers, also the variants using SMP and GPU computing are being developed. In this project, Tomáš is the the person doing the majority of work. You can learn more here (will be available soon).

Level set methods for image segmentation

This project is of a little bit more theoretical nature; Jan has choosen this topic for his doctoral thesis. Nevertheless, we all do regard the variational (and level set) methods as very interesting and promising for processing the images, for example, for image segmentation. Clear mathematical foundation belongs among the main pros. We have developed several improvements of the level set method for segmenting the images. Also the GPU implemenation has been carried out by Jan. Our main objective is to incorporate the shape of objects that are to be segmented into the method. You can learn more here (will be available soon).

Capturing images using in-car cameras

This project was done for one big European car maker. The main idea is that the cameras built into cars can also be used for capturing the images that are useful in traffic information systems. We have developed a hardware on-board unit receiving the images from the camera and transmitting them via WLAN or UMTS/GPRS to a portal application running on a server. The software for the on-board unit as well as the portal application have also been developed. By making use of our system, the images showing the traffic situation are available almost in real time for various purposes. You can learn more here (will be available soon).

Counting the objects in traffic

In this project, we are developing the methods for counting the moving objects, namely the cars, in the videosequences captured by a camera. The problem is of great practical importance in systems for traffic management since the traffic load that can be measured in this way is an essential input information. Thanks to OvaNet company, we have a camera installed on a busy street in the city of Ostrava. The sequences obtained from this camera are an important experimental terrain for us. You can learn more here (will be available soon).

Measuring surface tension of melted metals and slags

Surface tension is an important physical property of liquids. In industrial environment, it is also important in case of melted metals and slag. In this case, it is quite difficult to measure it. We have developed the software for measuring the surface tension by analysing an image of a drop of melted material. This activity is a part of the project that was funded by the Grant Agency of The Czech Republic. You can learn more here (will be available soon).

Intelligent camera

We have developed an intelligent camera, which is a camera that contains a computer in its body. Its computational power is sufficient to solve the majority of image processing and analysis tasks directly inside the camera. Therefore, there is no need to transmit the images from the camera. Only the results of the analysis can be transmitted, which reduces the amount of data dramatically. We have manufactured several prototypes of the camera that are used both in indoor as well as in outdoor conditions. You can learn more here (will be available soon).

Automatic car license numbers reading

We have developed a fast and reliable module for reading the car license numbers. The module is now incorporated in bigger systems by a company and by our lab too. The image to the left shows mounting an intelligent camera equipped with our system for reading the license numbers in the underground parking lot of our university. You can learn more about this project here (will be available soon).

Mobile robot for 3D scene reconstruction

We have created a software for a mobile robot the purpose of which is to reconstruct the scene in the neighborhood of robot. The robot is equipped with a pair of CCD cameras and is controlled remotely. It was intended to use at places which the human should not enter. The image shows the whole robot. Its mechanical part was carried out by the Faculty of Mechanical Engineering; testing was done by the Faculty of Coal Mining and Geology. This activity was a part of project that was funded by the Grant Agency of The Czech Republic. You can learn more here (will be available soon).

Corner detection algorithm

cordet The corners and vertices are important features in images, which are frequently used for scene analysis, stereo matching and object tracking. Ma
  • Milan Šurkala
  • ny algorithms for detecting the corners have been developed up to now. We have developed a new algorithm with good properties in a certain sense. You can learn more about this project here.

    RFID Emulator

    cordet At present, RFID cards are widely used, which may be a security risk. The emulator we have developed demonstrates that problem. It can emulate any RFID card. For more details, please, contact Michal, who did that work. We do not present the details here intentionally.