Currently, detecting and counting bicycles is achieved by detecting changes in air pressure in pneumatic tubes across a road over which these bicycles pass. Despite this method being relatively cheap, this method is not optimally accurate when deployed across multiple lanes or when deployed on lanes where overtaking is prevalent, insights other than throughput - such as size, velocity or cycling direction cannot be measured and lastly these tubes and other machinery are inherently exposed to wear and tear due to physical contact and moving parts (Brosnan et al.,2015). For this reason, the company MINDHASH has requested us to look into the feasability of using a Solid State LiDAR sensor (cheaper than rotational LiDAR with a more limited FOV) as a way to provide an affordable and more accurate way to detect and count bicycle lane traffic. In this project, we propose and display the feasibility of a method by which it is possible to detect and classify bicycles with a solid state LiDAR module, and display insights into this traffic data in a visually attractive and interactive fashion. As part of this, we have created a point cloud data collection tool for the Livox Mid-40 Solid State LiDAR unit, which could be used by researchers around the world for data collection, created a valuable and unique dataset containing 620 labelled data frames, each consisting of approximately 50.000 data points, in which bounding boxes have been drawn for all bicycles in a frame, created a machine learning model with high AOS and mAP (higher than those reported in the original paper of the approach used (Lang et al., 2019)) to precisely detect and count bicycles in a given point cloud frame and lastly created a web application providing valuable insights into the detected bicycle lane traffic data.
The CS Design Project module is one of the two final modules of the Bachelor. In the design component of this module, students show that they master the entire design trajectory, from the first informal specification of requirements by a client to the delivery and presentation of a well-documented working product. Projects are submitted by clients from either inside or outside the University. Students perform the project in groups of 3-5 students under the supervision of a teacher from the Department of Computer Science. The supervisor is also the one who assesses the process and products of the group. Project deliverables include a project proposal, a design report, a presentation and a poster.
The design report for this project.
View the full design report for this project.
Currently, detecting and counting bicycles is achieved by detecting changes in air pressure in pneumatic tubes across a road over which these bicycles pass. Despite this method being relatively cheap, this method is not optimally accurate when deployed across multiple lanes or when deployed on lanes where overtaking is prevalent, insights other than throughput - such as size, velocity or cycling direction cannot be measured and lastly these tubes and other machinery are inherently exposed to wear and tear due to physical contact and moving parts (Brosnan et al.,2015). For this reason, the company MINDHASH has requested us to look into the feasability of using a Solid State LiDAR sensor (cheaper than rotational LiDAR with a more limited FOV) as a way to provide an affordable and more accurate way to detect and count bicycle lane traffic. In this project, we propose and display the feasibility of a method by which it is possible to detect and classify bicycles with a solid state LiDAR module, and display insights into this traffic data in a visually attractive and interactive fashion. As part of this, we have created a point cloud data collection tool for the Livox Mid-40 Solid State LiDAR unit, which could be used by researchers around the world for data collection, created a valuable and unique dataset containing 620 labelled data frames, each consisting of approximately 50.000 data points, in which bounding boxes have been drawn for all bicycles in a frame, created a machine learning model with high AOS and mAP (higher than those reported in the original paper of the approach used (Lang et al., 2019)) to precisely detect and count bicycles in a given point cloud frame and lastly created a web application providing valuable insights into the detected bicycle lane traffic data.
The CS Design Project module is one of the two final modules of the Bachelor. In the design component of this module, students show that they master the entire design trajectory, from the first informal specification of requirements by a client to the delivery and presentation of a well-documented working product. Projects are submitted by clients from either inside or outside the University. Students perform the project in groups of 3-5 students under the supervision of a teacher from the Department of Computer Science. The supervisor is also the one who assesses the process and products of the group. Project deliverables include a project proposal, a design report, a presentation and a poster.
The design report for this project.
View the full design report for this project.