This sample project focuses on “Vechicle Detection, Tracking and Counting” using TensorFlow Object Counting API.
The TensorFlow Object Counting API is used as a base for object counting on this project, more info can be found on this repo.
The developing is on progress! This sample project will be updated soon, the more talented traffic analyzer app will be available in this repo!
General Capabilities of This Sample Project
This sample project has more than just counting vehicles, here are the additional capabilities of it:
- Detection and classification of the vehicles (car, truck, bicycle, motorcycle, bus)
- Recognition of approximate vehicle color
- Detection of vehicle direction of travel
- Prediction the speed of the vehicle
- Prediction of approximate vehicle size
- The images of detected vehicles are cropped from video frame and they are saved as new images under “detected_vehicles” folder path
- The program gives a .csv file as an output (traffic_measurement.csv) which includes “Vehicle Type/Size”, ” Vehicle Color”, ” Vehicle Movement Direction”, ” Vehicle Speed (km/h)” rows, after the end of the process for the source video file.
- More powerful detection models will be shared.
- Sample codes will be developed to process different types of input videos (for different types of road traffics such as two way lane road).
- Code cleanup will be performed.
- UI will be developed.
The input video can be accessible by this link.
- Vehicle detection and classification have been developed using TensorFlow Object Detection API, see for more info.
- Vehicle speed prediction has been developed using OpenCV via image pixel manipulation and calculation, see for more info.
- Vehicle color prediction has been developed using OpenCV via K-Nearest Neighbors Machine Learning Classification Algorithm is Trained Color Histogram Features, see for more info.
TensorFlow™ is an open source software library for numerical computation using data flow graphs. Nodes in the graph represent mathematical operations, while the graph edges represent the multidimensional data arrays (tensors) communicated between them.
OpenCV (Open Source Computer Vision Library) is an open source computer vision and machine learning software library. OpenCV was built to provide a common infrastructure for computer vision applications and to accelerate the use of machine perception in the commercial products.
Source video is read frame by frame with OpenCV. Each frames is processed by “SSD with Mobilenet” model is developed on TensorFlow. This is a loop that continue working till reaching end of the video. The main pipeline of the tracker is given at the above Figure.
By default I use an “SSD with Mobilenet” model in this project. You can find more information about SSD in here. See the detection model zoo for a list of other models that can be run out-of-the-box with varying speeds and accuracies.
The minimum vehicle detection threshold can be set in this line in terms of percentage. The default minimum vehicle detecion threshold is 0.5!
Demo video of the project is available on My YouTube Channel.
Docker setup with Nvidia GPU: Run the demo in