18464257. CONTROL APPARATUS, CONTROL METHOD, AND PROGRAM simplified abstract (FUJIFILM Corporation)

From WikiPatents
Jump to navigation Jump to search

CONTROL APPARATUS, CONTROL METHOD, AND PROGRAM

Organization Name

FUJIFILM Corporation

Inventor(s)

Tomoharu Shimada of Saitama-shi (JP)

Masahiko Sugimoto of Saitama-shi (JP)

Tetsuya Fujikawa of Saitama-shi (JP)

CONTROL APPARATUS, CONTROL METHOD, AND PROGRAM - A simplified explanation of the abstract

This abstract first appeared for US patent application 18464257 titled 'CONTROL APPARATUS, CONTROL METHOD, AND PROGRAM

Simplified Explanation

The patent application describes a control apparatus for a surveillance camera that uses machine learning to detect objects and adjust the imaging range accordingly. Here are the key points:

  • The control apparatus includes a processor that controls a surveillance camera.
  • It enables switching between a first surveillance mode and a second surveillance mode.
  • In the first mode, the camera acquires a first captured image and the imaging range can be changed based on instructions.
  • In the second mode, the camera acquires a second captured image and a trained model is used to detect objects in this image.
  • The imaging range is adjusted based on the detection result from the trained model.
  • The first captured image acquired in the first mode is used as a teacher image for machine learning.

Potential applications of this technology:

  • Surveillance systems in public places, such as airports, train stations, and shopping malls.
  • Security systems in residential and commercial buildings.
  • Traffic monitoring and management systems.
  • Wildlife monitoring and conservation efforts.

Problems solved by this technology:

  • Efficient monitoring of large areas by automatically adjusting the imaging range based on detected objects.
  • Improved accuracy in object detection through machine learning.
  • Reduction in false alarms and unnecessary alerts by using a trained model.

Benefits of this technology:

  • Enhanced security and safety by effectively monitoring and detecting objects in real-time.
  • Reduction in human effort and resources required for surveillance.
  • Improved efficiency and accuracy in object detection.
  • Potential for continuous learning and improvement through the use of teacher images for machine learning.


Original Abstract Submitted

A control apparatus includes a processor that controls a surveillance camera. The processor enables switching between a first surveillance mode in which the surveillance camera is caused to perform imaging to acquire a first captured image and an imaging range is changed according to a given instruction, and a second surveillance mode in which the surveillance camera is caused to perform imaging to acquire a second captured image, a trained model that has been trained through machine learning is used to detect an object that appears in the second captured image, and the imaging range is changed according to a detection result, and outputs the first captured image acquired in the first surveillance mode as a teacher image for the machine learning.