Embedded vision has been a deciding catalyst in the popularity of Automatic License Plate Recognition (ALPR) technology, transforming it from a niche application into a widespread tool for traffic management, security, and automation. This also made it feasible to deploy ALPR in a wider range of environments, from mobile units in law enforcement vehicles to fixed installations at high-speed toll booths, border control checkpoints, and parking facilities.
One of the key advantages brought by embedded vision to ALPR technology is the ability to process images on the edge, directly within the camera system. This local processing capability reduces latency, ensuring real-time identification and decision-making. It also minimizes the bandwidth needed for transmitting video streams to a central server, which enables the deployment of ALPR systems in locations with limited network connectivity.
Embedded vision has also made it possible to use advanced image processing and Machine Learning algorithms in ALPR systems. This has improved the accuracy of license plate detection and recognition across a variety of conditions, including varying lighting, weather, and vehicle speeds.
In this blog, you’ll get expert insights on the crucial features provided by e-CAM56_CUOAGX, e-con Systems’ latest camera. These features can be crucial for ALPR applications.
Features offered by e-CAM56_CUOAGX for ALPR applications
High sensitivity
In environments where lighting conditions can vary widely, from bright midday sunlight to low light at night or adverse weather conditions, e-CAM56_CUOAGX’s high sensitivity ensures that ALPR applications can capture clear, usable images across all scenarios. This is extremely important, as these applications rely on the accurate capture of plate numbers, which may be compromised by shadows, glare, or insufficient lighting.
High-sensitivity sensors allow for greater detail capture in low-light conditions without requiring additional illumination sources, reducing the system’s overall power consumption and enabling more discreet installation options. Furthermore, the ability to process these high-quality images quickly means that ALPR systems can operate smoothly in real time.
High resolution
The high resolution feature of e-CAM56_CUOAGX enables the capture of images with a greater level of detail, which is critical for accurately identifying and processing license plate numbers, especially from a distance or at high speeds. With high-resolution cameras, ALPR systems can distinguish fine details and small characters on license plates, thereby differentiating similar characters and improving recognition accuracy.
Additionally, high resolution image capture supports the recognition of plates from various regions, which may have different formats and sizes. In addition, high resolution allows for a wider field of view while maintaining enough detail for plate recognition, enabling a single camera to monitor multiple lanes of traffic or a larger area. It also reduces the need for additional cameras, simplifying installation and maintenance.
Finally, high-resolution images provide more data for algorithms to analyze, enhancing the accuracy of character recognition and reducing false positives. It can be important for ALPR applications that depend on accurate vehicle identification, such as toll collection, access control, or traffic law enforcement systems.
Global shutter mode
In rolling shutter cameras, images are captured line-by-line, which can cause distortion in fast-moving objects. e-CAM56_CUOAGX’s global shutter mode helps capture the entire image simultaneously. So, the images of license plates are free from motion blur and distortion, regardless of the vehicle’s speed or the camera’s movement.
With the global shutter mode, ALPR systems can accurately read plates in dynamic environments, such as highways or busy urban streets, where vehicles may be moving rapidly or changing lanes. Also, this technology is invaluable in low-light conditions or at night, where motion blur is more pronounced, guaranteeing consistent performance around the clock.
Sony® Pregius S IMX568 sensor
e-CAM56_CUOAGX is powered by the Sony® Pregius S IMX568 sensor, which is equipped with back-side illuminated (BSI) technology. This technology enables exceptional light collection efficiency, which can be a game-changer since ALPR applications operate across diverse lighting conditions, from bright daylight to low-light or night-time scenarios.
Additionally, the IMX568 sensor supports a wide dynamic range, enabling it to capture clear images in scenes with high contrast or sudden changes in lighting – a common occurrence in outdoor environments where ALPR systems are frequently deployed. Such adaptability enhances the system’s reliability, ensuring consistent performance under varying environmental conditions.
Speed detection test with e-CAM56_CUOAGX
e-con Systems conducted a speed detection test using the e-CAM56_CUOAGX, demonstrating the camera’s capability. The setup included a narrow-angle lens optimized for speed detection tasks affixed at a specific pole length and angle to accurately capture vehicles moving on the road below. The camera’s positioning was calculated to ensure a clear, unobstructed view of two reference points, R1 and R2, along the road.
To detect the car’s position relative to these reference points, e-con Systems employed YOLO V9, a cutting-edge object detection algorithm. It empowered the team to determine the speed of a vehicle by analyzing how quickly it moved between R1 and R2 based on the data extracted from consecutive frames. Following the initial detection, the DeepSORT algorithm was utilized to track the vehicle across multiple frames. DeepSORT’s ability to assign and maintain unique IDs for each vehicle ensured that tracking was uninterrupted, even in sequences where the vehicle might otherwise be lost due to occlusion or rapid movement.
The next phase of the test involved calculating the distance covered by the vehicle in a given timeframe, using the frame rate (FPS) to measure the device’s processing power. The data was then compared to pre-calibrated real-time data, establishing a benchmark for accuracy. The final speed calculation of each vehicle was derived using the formula Speed = Distance / Time.
In a nutshell, e-CAM56_CUOAGX’s global shutter mode helped capture crisp images of vehicles frame by frame without motion blur, ensuring accurate speed detection and tracking.
e-CAM56_CUOAGX is e-con Systems’ newest camera for ALPR applications
e-CAM56_CUOAGX, a 5-megapixel MIPI CSI-2 multi-camera system, utilizes a 1/1.8″ Sony Pregius S IMX568 sensor with back-side illumination for superior image acquisition. It is compatible with the NVIDIA® Jetson AGX Orin™ development kit, offering superior uncompressed YUV format image quality, further improved by its fixed focus and high sensitivity features.
Featuring a global shutter mechanism, the e-CAM56_CUOAGX captures sharp images of rapidly moving objects without distortion, making it suitable for use cases such as ALPR. It also supports the connection of up to six camera modules, providing flexibility with a standard M12 lens mount that accommodates a range of custom optics or lens options.
Suresh Madhu is the product marketing manager with 16+ years of experience in embedded product design, technical architecture, SOM product design, camera solutions, and product development. He has played an integral part in helping many customers build their products by integrating the right vision technology into them.