The thesis deals with the problem of synchronization of video signals and signals of different sensors. First, I investigated the possibilities of synchronized capture and storage of sensor signals and signals from two different video cameras: webcams and GoPro cameras. Both types of cameras are very widely available: the webcam is now installed almost in every personal laptop, and the GoPro camera is most often used for recording various outdoor activities. The GoPro camera is a completely autonomous device, webcam however only works in conjunction with the computer. In the capture application, the video signal from the webcam can be recorded in synchronization with the signals of other sensors. This could not be achieved with the GoPro camera, as the amount of data is too large for real-time processing. With the GoPro camera, I therefore used the option of local storage in the camera's memory with precise time management and placement of time tags. The application on the PC manages the operation of the GoPro camera using the control messages it sends through a wireless Wi-Fi connection.
The introduction of the thesis includes a description of the importance of the correct technique of a sportsman in sport, and consequently the use of technology in achieving better results. Some products from the market are also shown, but they have some shortcomings or could be slightly upgraded. For this reason, an idea has been created for making a system that allows for precise analysis by combining video analysis and analysis with sensor signals, and the system must also allow simultaneous synchronization of capture and storage of signals from wireless motion sensors and image signals from the camera, and provide an interactive display of recorded signals for the purpose of analyzing movement in sports.
The second section gives the design of a system for capturing, storing and displaying sensor signals. The hardware of the measuring system consists of a laptop with a USB webcam, a wireless motion sensor with built-in accelerometer and gyroscope, a GoPro camera and a wireless router. The connection between the PC and the GoPro camera and between PC and the motion sensor are carried out separately through two different Wi-Fi interfaces of the PC.
The third chapter provides a description of the hardware and software of the planned system. In particular, the technical characteristics of video cameras are described first, followed by a description of the performed measurements when using different video players and a more detailed description of the VLC video player I used. Cameras differ primarily in image quality, and the result is the large differences in the amount of data captured. The differences are also in the video format. In the third chapter, I therefore focused on dealing with various video file formats and various video encoders. A description of the program code of the autonomous sensor device, which is produced in the Arduino IDE environment, is also given. The main application runs on a PC and is made in the LabVIEW programming environment.
Chapter four describes the operation of designed applications for capturing, viewing and analyzing sensor signals and video signals from both cameras. Instructions for using all the applications describing settings of different parameters are given. The system is built modularly and therefore four applications are presented separately. The first application controls the operation of the webcam and the GoPro camera and performs synchronized capture of all signals and storage on a personal computer disk. Second application is designed for viewing and analyzing recorded sensor signals and video clips of the webcam. This works in a mode that allows stopping and playback with adjustable speed. The next application is intended for synchronized display of sensor signals along with the GoPro camera video footage. Another application is described Separately, for capturing images from clips of GoPro camera at selected times, with resolution that we set in the first application. The conclusion of the fourth chapter also provides a practical test of the operation of the designed system for synchronized capture, display and analysis of sensor signals.
The thesis in conclusion summarizes the main findings and ideas for continuing work. The system can be expanded with additional sensors, and application modules for capturing and displaying recorded signals can already be used in applications to help with the learning of motion in sports.