TECHNOLOGY

문서위치

  • Home
  • Technology
  • Intelligent CCTV

Intelligent CCTV

Object Recognition
Approaches for object detection are generally classified as either machine learning based access or deep learning based access.
Features
Approaches for object detection are generally classified as either machine learning based access or deep learning based access. For machine learning based access, It first needs to define using one of the methods below and then classify using techniques such as the Support Vector Machine (SVM). Meanwhile, Deep learning techniques can detect end-to-end objects without specifically defining their functionality and are usually based on the convolutional neural network (CNN).
Uses
Computer technology, which is related to computer vision and image processing, deals with the detection of specific classes of semantic object instances (eg, humans, buildings, cars) in digital images and video.
Well-researched object detection areas include face detection and pedestrian detection.
Object detection has been applied to a number of computer vision applications, including video recovery and video surveillance.
Face Recognition
It is a technology of extracting facial feature points and comparing it with the data in the database to verify identity.
Features
It is a technology of extracting facial feature points and comparing it with the data in the database to verify identity. The method of extracting facial feature points is as follows.
- Thermal imaging method using infrared camera that recognizes heat spots in blood flow through arteries and veins
- Identifying facial features such as angles and distances between eyes, mouth, nostrils, jaws, and bone prominence using a three-dimensional measuring instrument
Uses
Recently, as the accuracy of facial recognition is increased by using artificial intelligence (AI) technology such as deep learning, application cases are rapidly increasing in various fields such as public, access control, finance, and distribution. In the future, to apply face recognition-based identity identification services, information subjects should participate in the entire process of collecting, storing, using, and discarding face recognition information that is distributed. Also, measures should be in place to safely manage spills.
Object tracking
It is a technique for estimating the change of an object by using the similarity between characteristic information such as size, color, shape, outline,
Features
It is a technique for estimating the change of an object by using the similarity between characteristic information such as size, color, shape, outline, etc. of an object in a series of image frames. In other words, if the path of interest in an image is found to be similar to the previous frame, and then recognized as the same object, it can be said to keep tracking the object. Object tracking methods include Point Tracking, Kernel Tracking, and Silhouette.
Uses
It is used in various fields to find a change in the position of a specific object such as a person, an animal, or a vehicle in an image captured by a camera. In particular, it is actively used in real-time video security, video call, traffic control, augmented reality.