Machine vision camera technology is no longer limited to production process, observation and inspection but is increasingly applied to applications analyzing athletes and sporting events. There has been a rise in demand for camera technology across a variety of sports, including athletics, baseball, basketball and golf.

The Industrial machine vision cameras from GeT Cameras are capable of detecting certain movements more accurately than the human eye and can bring these movements into sharper focus. However, determining the best machine vision setup for sports analysis can be challenging. That is why the machine vision experts team of GeT Cameras provides this step-by-step guide to selecting the appropriate camera, lens, lighting and software for this application.

Machine vision cameras for sports analysis

The client had prior experience with camera technology in a previous setup, where they used a low-resolution monochrome GigE camera with a frame rate of 120 frames per second. However, the image quality was insufficient for their needs, prompting them to look for a higher-quality camera setup. The client mentioned using a low-resolution camera at 120 fps previously, so the new setup was required to have a higher frame rate, with a minimum resolution.

After conducting some preliminary research, the client shortlisted a few cameras using a user-friendly website. Another consideration was the movement of the camera during image capture, for which a global shutter camera was recommended. More details on the difference between global and rolling shutter cameras can be found in an article from the knowledge center titled “Rolling Shutter vs Global Shutter.”

In the previous setup, GeT Cameras’ client used a monochrome GigE camera to analyze athletic events, which met their needs. For his new set up he made some adjustments and decided to go for a USB3 camera instead of GigE. With a few changes in his set up he was able to bring his PC closer to the camera allowing a shorter cable length. This meant that he could take advantage of the extra framerate that the USB3 cameras have to offer. He also expressed interest in using colour images this time. As there was no price difference between colour and monochrome cameras, the decision was left to the client. After reviewing the options, the client was torn between the 0.4 MP camera ‘MER2-041-528U3C‘and the 1.6 MP camera ‘MER2-160-227U3C‘ . Due to concerns about resolution limitations with the 0.4 MP model, it was advised to opt for the 1.6 MP camera, which offered better resolution with only a minimal price difference.

Analyzing athletics with a lens for the IMX273 sensor

The selected ‘MER2-160-227U3C’ camera is equipped with a Sony IMX273 sensor with a 1/2.9” sensor size. To choose the right lens you can make use of a lens calculator. This tool, available on the website and allows users to input the camera resolution,  desired field of view and working distance to calculate the appropriate focal length for the lens.

The client’s setup had a fixed working distance of 11 meters and they aimed for a field of view of at least 6 by 5 meters. After consultation, the field of view was set to 6.8 meters, resulting in an 8mm focal length. Based on this focal length, a suitable option was identified: the ‘LCM-5MP-08MM-F1.4-1.5-ND1.’ This 5 megapixel 8 mm lens  lens pairs well with the 1.6 MP camera ‘MER2-160-227U3C‘ for capturing high-quality images in various sports. The low F-number offers a great light sensitivity for low lighting situations.

Machine vision lighting for sports analysis

For this project, the client indicated that they did not intend to use machine vision lighting, as all images would be captured in daylight. However, for similar applications in low-light conditions, the use of lighting is recommended. A suitable option is the NIR LED spotlight, Industrial LED Spot light NIR 940nM,’ which operates in the 940NM range and offers 28 watts of power. If the camera and lens setup have a horizontal opening angle greater than 20 degrees, it is suggested to use at least two LED spotlights for more even lighting.

For additional guidance on selecting the best lighting for similar applications, contacting technical support from GeT Cameras is recommended.

Image processing software for sports inspection

A complete machine vision system for sports analysis requires compatible computer vision software. To control the industrial cameras, the installation of an SDK is necessary, which can be downloaded for free from the company’s download page.

In this case, the client approached the company on the recommendation of Kinovea, an open-source software designed for human movement analysis. Kinovea is ideal for image capture, observation and measurements and the cameras are compatible with Kinovea version 0.93 or higher.

For more details on using machine vision cameras with Kinovea, an article in the knowledge center titled “How to use Kinovea with machine vision cameras.” offers additional information.

Expert support for sports analysis camera systems

Machine vision technology surpasses the capabilities of the human eye by capturing precise movements in various sports, including athletics, baseball and basketball, at high speeds and resolutions. This enables sports analysts to gain deeper insights into athletic performance, which can lead to improved training methods and strategic decision-making.

For those seeking assistance in creating a machine vision system for sports analysis, expert support is available. Interested parties are encouraged to contact GeT Cameras for guidance on finding the ideal setup.