Presentation of the Prexable Software for Video Detection in SRC’s Surveillance Cameras
Presentation of the Prexable Software for Video Detection in SRC’s Surveillance Cameras
In a recent session with the Sadjad Research Center (SRC), our team presented the latest developments in Prexable, our cutting-edge software for video detection. This software is designed to elevate SRC’s surveillance capabilities by accurately identifying individuals and objects of interest, enabling enhanced security, and offering comprehensive video insights. Below is a summary of the items demonstrated, including our team’s responses to SRC’s specific requests.
Summary of Presented Items
1. Customizable Output Process Demo
Video Demo Using Client Data: To showcase Prexable’s adaptability, we used SRC’s own surveillance footage, providing a realistic view of the software’s performance in a real-world environment.
Individual Detection with Unique ID Assignment: Prexable efficiently detects individuals within each frame, assigning them unique IDs. These IDs track each individual’s timeline, visible at the bottom of the video, allowing SRC to easily monitor movement and duration.
Parameter Customization: One standout feature of Prexable is its customization flexibility. Every parameter in the video detection process can be tailored according to SRC’s needs, including sensitivity adjustments, detection frequency, and object categories.
Presenting Detection Results: We provided SRC with a live demonstration of the detection results, showcasing the clarity and accuracy of the outputs.
2. Advanced Detection Capabilities
Detection of Individuals and Key Objects: The system can accurately detect individuals and select objects of interest, enabling SRC to focus on relevant data.
Optimal Performance with YOLOv8: Leveraging the YOLOv8 algorithm, Prexable achieves fast, precise object detection suited for SRC’s surveillance needs.
3. Privacy Options and Server Deployment
Client Server Option: For enhanced privacy, Prexable can be deployed directly on SRC’s servers, ensuring that all data remains within SRC’s infrastructure and is not uploaded externally.
4. Status of Detection and Recognition Algorithms
Detection: The detection component is fully functional, successfully identifying and tracking individuals.
Recognition: Our team is currently developing the recognition component. While IDs are assigned to individuals, recognition is not yet fully implemented. This means that an individual may receive a new ID if they leave and re-enter the frame.
Responses to SRC’s Requests
1. Recognizing Authorized Individuals for Easy Identification of Unauthorized Personnel
Status: In progress. Our team is actively working on enabling the recognition of authorized individuals, which will streamline the identification of any unauthorized entries.
2. Nighttime Detection
Current Capability: Prexable already supports nighttime detection with good accuracy.
Enhancement: We are further enhancing the algorithm’s sensitivity to optimize detection under low-light conditions.
3. Interactive Timeline for Video Analysis
Feature Addition: Upon clicking on an individual’s timeline bar, SRC will soon be able to view specific timestamps of their presence. This functionality will offer SRC a more interactive way to review footage.
4. Position Tracking of Authorized Individuals
Status: Under development. This feature will eventually allow SRC to not only identify but also track the location of authorized individuals throughout the footage.
5. Full On-Site Server Deployment
Upcoming Solution: The private version of Prexable will soon be deployable directly on SRC’s server. This option, coupled with SRC’s own server infrastructure, will ensure maximum privacy and eliminate any need for external data uploads.
The presentation of Prexable highlighted its customizable and powerful video detection capabilities. SRC’s feedback and requests are invaluable as we refine and optimize features for practical surveillance solutions. As we continue to develop and implement these features, Prexable promises to become a pivotal asset in SRC’s surveillance ecosystem, providing robust detection, privacy-focused deployment, and adaptable insights tailored to SRC’s needs.
