<img height="1" width="1" style="display:none" src="https://www.facebook.com/tr?id=159683641205248&amp;ev=PageView&amp;noscript=1">

IP Cameras with Built-in Machine Learning AI Capabilities

Matt Golueke June 17, 2019 20:46 PM

Bosch Camera Trainer video thumbnail- Trinklein

 

Today marked our official introduction to Camera Trainer machine learning, the next level of AI-enabled video analytics available in select Bosch IP cameras for real-time detection and alarm alerting. The data created and stored by the Bosch camera, or metadata, can be used with data collection systems to create extensive reports and statistics. 

Think of Camera Trainer as the third and highest level of capability within the Bosch security family of Narrow AI (task-specific artificial intelligence) within video analytics for specific camera models within the CPP6, CPP7, and CPP7.3 families:

  • Essential Video Analytics
  • Intelligent Video Analytics
  • Camera Trainer - Machine Learning

A properly trained technician or experienced end user with training can use prior video footage to help identify specific objects in a scene in a variety of lighting and environmental conditions by providing positive and negative feedback to the camera when presented with objects to identify. 

This information is then used to provide much higher reliability for detection and alerting for given behavior rules.

  • Step 1: register for Bosch Security Academy here >>
  • Step 2: register for the Camera Trainer free online class - (45 min) here >>
  • Step 3: request a license key, available to people who have taken and completed the short online training class here >>

With Camera Trainer it is possible to go beyond merely the detection of pre-defined objects. For example, specific Bosch video cameras can be taught to recognize people with a shopping cart.

Another advantage of the new machine learning-enabled video analytics feature is that it is now feasible to reliably separate vehicles clustered together in front of traffic lights, regardless of movement. Or determine how long a vehicle has been parked in a parking spot. And determine if parking spots are either full or empty. With Camera Trainer, an object no longer needs to move to create a change in a situation, because the camera can be trained to also analyze stationary situations and compare them over different points in time.

Now, imagine the benefits of translating all of this insightful information into instant alerts and business operations intelligence. You could be optimizing your business processes or improving customer satisfaction, with maximum security and safety awareness for your people, customers, and property.

In the example video below, we see the results of programming a Bosch camera to recognize the presence of a specific sign in the field of view by using machine learning in Camera Trainer:

 

DEMONSTRATION: contact us to set up a personal demonstration >>

demobosch

 

Compatible Camera List (rev 5/2019) here >>

Camera Trainer compatible camera list image - rev5-2019

 

Camera Trainer Technical Note here >>

Camera Trainer tech note image for v6.20

 

Watch videos about Camera Trainer setup here >>

Camera Trainer Playlist image

 

More about Bosch IP camera machine learning here >>

video_analytics_machine_learning_res_1280x720

 

More about customized applications here >>

video_analytics_customized_applications_res_1280x720

 

Camera Trainer use cases here >>

video_analytics_camera_trainer_use_cases_res_1280x720

 

Setting up Camera Trainer here >>

video_analytics_ease_of_use_and_installation_res_1280x720

 


 

Video Analytics example videos image

 

Video analytics lens calculator image


 

Video analytics echnical documentation image

 

DEMONSTRATION: contact us to set up a personal demonstration >>

demobosch

 

 

 

 

Topics: Technical - Simplified, Proof of Concept, Your Challenge, Our Solution

View More Blog Topics

Trends, interviews, and practical advice on all things Security - Return to Main Blog Page

Search Our Blog

Subscribe for Blog Updates