Image analysis system provides coaching data for cyclists

Feb. 4, 2020
Better posture will provide more comfort and better performance.

Cycling, a popular international sport and hobby, puts a lot of stress on the body. Ten hours of cycling equates to approximately 54,000 pedal strokes and as many bends of the knee, hip, postural muscles, as well as joint restrictions. Such movements put great demands on tendons and joints, which can lead to dysfunctional movements, pain, and injuries. A 2010 study, Overuse Injuries in Professional Road Cyclists ( shows that professional cyclists often complain of pain, especially in the back (85%) or in the knees (57%).

French cycling analysis and coaching company AR-Entraînement (La Pallet, France; focuses on performance requirements and physical abilities of individual cyclists, since each person has different requirements in terms of physical size and ability. In collaboration with the professional racing bike team Vital Concept B&B Hotels, AR-Entraînement has developed an image analysis system used in the company’s performance center that helps provide data for subsequent analysis and coaching.

The company, explains Managing Director Alban Renaud, leverages real-world experience and input from fitness trainers and osteopathic physicians from the world of cycling. For analysis, a bicycle is fixed in a roller trainer and a UI-3240LE USB 3.0 camera from IDS Imaging Development Systems (Obersulm, Germany; is positioned perpendicular to the bike and rider to record in a 2D X-Y axis. The UI-3240LE is based on a color EV76C560ACT CMOS image sensor from Teledyne e2v (Chelmsford, Essex, UK;, which features a 5.3 µm pixel size and reaches frame rates of up to 60 fps.

When the cyclist pedals with varying intensity and in different positions, the camera captures the motion sequence and records changes in joint angles at 60 fps. The camera is attached via USB 3.0 cable to a laptop running the open-source Kinovea sports analysis software (

Direct effects on mechanical parameters are measured, such as the influence of force, movement, and pedal technology. Movement-related deformations are factored in, since pedal movement is not simply pedalling in a circle and joint angles of riders show deviations of up to 15°.

The software allows manual measurement of angles, distances, and times, as well as tracking to trace the trajectories of points in the video. It also allows slowing down of frame rate for highlighting certain movements, image enlargement up to 6x, and the measurement of alignment or connection angles on selected images to evaluate the need to make adjustments on the bike.

Changes in joint angles during pedaling are visualized in a diagram within the software, with the trajectory of markers positioned on the cyclist serving as the basis. Furthermore, the software can display multiple videos at once to observe the cyclist in different shots to compare settings like sitting position, handlebar position, saddle height, and the angle at which the feet touch the pedals.

A number of considerations led to AR-Entraînement choosing the IDS Imaging Development Systems industrial camera for the image analysis system. First, the camera offers a global shutter CMOS image sensor, which helps to ensure that captured images are not distorted during movement, says Renaud. At 47 x 46 x 26.3 mm, the camera is also easy to transport, since the system must be mobile. Additionally, the camera offers four areas of interest (AOI), which allows several characteristics to be checked simultaneously or allows the AOIs to be captured in an exposure series with different parameters.

Cyclists themselves may lack biomechanical and kinematic relationship knowledge, so with this system, AR-Entraînement aims to provide this information to assist cyclists in their hobby and health. Renaud believes that, with better aligned forces, more useful mechanical energy is generated with the same energy input, meaning the same effort but higher speeds.

In addition, he believes that a better posture will provide more comfort and better performance, enabling riders to ultimately “position themselves optimally on the bike to prevent pain, minimize the risk of injury, and maximize sporting success.”

Related stories:

4D tracking system recognizes the actions of dozens of people simultaneously in real time

High-speed cameras help visualize trajectories of flying insects

Share your vision-related news by contacting Dennis Scimeca, Associate Editor, Vision Systems Design


About the Author

James Carroll

Former VSD Editor James Carroll joined the team 2013.  Carroll covered machine vision and imaging from numerous angles, including application stories, industry news, market updates, and new products. In addition to writing and editing articles, Carroll managed the Innovators Awards program and webcasts.

Voice Your Opinion

To join the conversation, and become an exclusive member of Vision Systems Design, create an account today!