In recent years, smart phones on the market have undergone a huge developement in the field of hardware and software. Thus, the today's modern phones have become well suited to run computer vision algorithms on them. Taking advantage of this opportunity the possibility of implementation of innovative software opens, that are ingrained in people's everyday lives to make convenient carrying out their everyday tasks.
Such an example is a mobile software application that supports the buying process. If a customer takes a photo of the intended product with his phone, it recognizes the specific object and the application provides the customer with some information. Then the customer can even buy the product identified by the phone, thereby avoiding standing in the queue. An other opportunity is recognizing buildings and giving some information for the tourists. Moreover there is a possibility to detect a company's logo, and the application gives you more information about it's products, and the program can even navigate the user to the nearest shop or restaurant based on the GPS coordinates of the phone.
The aim of my thesis is to investigate the nowdays widespread detection algorithms that are suitable for supporting the buying process. I examine the tools of the Open Source Computer Vision Library (OpenCV) whether they fit to the task or not. From them I examine the cascade classifiers in details and I introduce the results of them.