Introduction
Smart vehicles and intelligent transportation system technologies are continuing to alter many aspects of human life. As a result, technologies like automatic license plate detection (ALPR) have become commonplace in our daily lives. Furthermore, the concept of ALPR has the potential to contribute to a variety of application cases while obviating the need for human participation.
VIDIZMO, in combination with its all-encompassing, robust video content management capabilities, has built AI Indexer to provide enterprises with the power to detect license plates along their numbers automatically within the videos, images and later redact them too.
How does VIDIZMO leverage ALPR?
VIDIZMO leverages ALPR technique by building a robust model, which in turn uses deep learning to detect license plates. The model reads license plates on automobiles automatically and quickly, without the need for human intervention.
VIDIZMO indexer takes videos, images as input. Indexer starts a workflow activity that takes some time to detect license plates in video. Video processing time may vary as it depends on factors such as video quality, duration, and resolution. Once the detection is complete, the user can see the detected license plates by navigating to Studio Space and redact all the detected license plates.
On the detected license plates VIDIZMO OCR is applied to detect the license plate numbers. As in case of video we apply rule based tracker to identify objects(in this case License Plate). For each detected License Plate we apply VIDIZMO OCR on all the frames where the respective license plate is present. For the assignment of license plate number the system looks for the number which is repeated in most of the frames.
Moreover, VIDIZMO Indexer also gives the flexibility to choose from three different models, which are "Small", "Medium", and "Large". It also allows the user to set confidence threshold, tracking frame based on which the indexer will detect license plates and their numbers so that the user is able to get most out of the VIDIZMO Indexer capability according to the use case.
Please refer to the following article to understand what model size, confidence threshold, and tracking frames are: How to configure VIDIZMO Indexer for object detection
In order to redact the automatically detected license plates, please refer to this article: How to Redact PII(Personally Identifiable Information) from Digital Evidences
Detection
VIDIZMO Indexer's Automatic License Plate Recognition capability offers numerous advantages that are the basis for real-world scenarios. The majority of its advantages are related to automating manual jobs, governance, and improving the customer experience by automatically detecting the license plates and their numbers within the video on a single click rather manually annotating all the license plate to later redact them.
To understand redaction in detail, please refer to this article: Understanding Redaction Using Studio Space Some of the advantages, VIDIZMO indexer offers for Automatic License Plate Recognition are listed below:
Advantages of VIDIZMO Indexer
VIDIZMO Indexer has been made such that no human input is required for precise and fast number plate detection. Moreover, it also minimizes detection job wait times by processing multiple jobs in parallel. As a result, it promotes cost-effective governance and shortens wait times.
VIDIZMO Indexer has been built keeping in mind the importance of detecting license plate in night time and as important it is detecting in day time. As a result, license plates in night time are detected as efficiently as daytime . As a result, it promotes higher detection accuracy and less manual work.
VIDIZMO Indexer has been built keeping in mind the importance of detecting license plate in every weather condition including sunny, cloudy, and rainy. As a result, it promotes higher detection accuracy and less manual work.
Highlights
Detection Accuracy
As of now, our VIDIZMO Indexer has been built to perform best on USA License Plate with MAP(Mean Average Precision) of over 87%.
Contributions were made by Nabeel Ali & Mustafa Zulfiqar.
Read Next:
Understanding Redaction Using Studio Space
How to Redact PII From Digital Evidences
Understanding Automatic Weapon Detection
Understanding Automatic Vehicle Detection
How to configure VIDIZMO Indexer for object detection