Asahito Mirai Technology Co., Ltd. will release the “AI Video Security 5-Stage Design Framework,” which organizes the basic concepts for proposing AI video security.
This framework not only introduces the functions of AI-enabled cameras and recorders,
- Define the on-site challenges
- Design the shooting conditions
- Select AI functions
- The entire system is composed of
- Organize the operation and verification methods.
This process involves five stages to transform on-site challenges into concrete video security systems that can actually be implemented and operated.
Also Read: Dai Nippon Printing Modernizes ESG Management
Our company has maintained ongoing business relationships with over 500 clients nationwide, and has been involved in supplying products and selecting equipment tailored to diverse facilities, installation environments, delivery dates, and operating conditions.
This time, we have organized the concepts we have emphasized into five stages and are publishing them along with two diagrams.
In recent years, AI-enabled cameras and recorders have been equipped with a variety of functions, including person and vehicle identification, intrusion detection, line crossing detection, video search, people counting, heat maps, automatic tracking, and audio and light warnings.
However, the fact that a product is equipped with AI functionality does not necessarily mean that that functionality will be useful in real-world situations.
For example, even with the same “intrusion detection” function, the detection results will vary depending on conditions such as the camera’s installation location, shooting angle, distance to the target, daytime/nighttime brightness, background movement, obstructions, and rain or snow.
Furthermore, even if AI detects an anomaly, if it’s not decided who will check the notification and under what conditions they will contact the site, that information cannot be translated into actual action.
Furthermore, it’s necessary to check system requirements other than AI functionality, such as the camera and recorder combination, recording storage period, HDD capacity, and network bandwidth.
SOURCE: PRTimes


