Tag Archives: detection

AI-Powered Motion Detection

8 Replies

In our macOS CCTV software SecuritySpy, deciding whether to trigger recording and actions (e.g. notifications) is a two-step process. Firstly, a motion-detection algorithm determines whether there is some interesting movement happening. Next, the image is cropped to the area of movement and passed to a neural-network classifier to determine whether there is an object of interest present, like a human, vehicle or animal.

The classifier in SecuritySpy is based on a ResNet-50 neural network, custom-trained on over 100,000 real CCTV images from thousands of different cameras. It is a highly optimised, powerful network, producing very accurate results (around 97% for humans and vehicles).

In contrast, until now, the motion detection algorithm has been fairly simple. It has worked via pixel-based comparisons between successive video frames, with some additional filtering to help ignore extraneous movement caused by noise, transient motion and lighting changes.

This standard motion detection is fast and very effective in controlled environments that don’t suffer from a lot of extraneous motion. However, it is less effective in more challenging outdoor environments, where movement caused by shadows, swaying plants, snow, rain and so on can cause a high number of false-positive detections. In such situations, it maintains high sensitivity (ability to detect true motion when it occurs), but suffers from poor specificity (ability to accurately ignore extraneous movement).

SecuritySpy’s classifier is able to filter out virtually all these false-positives, however even if only 3% of false-positive detections get through the classifier, in particularly challenging situations where the motion detector is generating large numbers of false-positives, inevitably there will still be a significant number getting through and causing triggers.

Standard Motion Detection Pipeline

Standard motion detection pipeline: with False Positives (FP) greatly outnumbering True Positives (TP), even though the classifier is very accurate, a significant number still get through and cause triggers

Continue reading →

How To Achieve Effective Motion Detection

12 Replies

SecuritySpy’s motion detection algorithm employs many techniques in order to accurately detect real motion events while minimising the rate of false-positive detections. But besides an effective algorithm, there are many choices about the setup and configuration of your video surveillance system that will help achieve reliable motion detection – these are outlined below.

1. Camera Angle

How you position and point the camera, as well as the focal length of the lens (how wide-angle it is) makes a big difference to the effectiveness of motion detection. The key points to consider are as follows:

  • The activity you want to capture should be relatively large in the frame. Don’t use a camera with a very wide-angle lens, as this will make objects and people appear very small – too small to trigger motion detection or make out any important features such as faces.
  • Point the camera downwards and don’t include any sky in the frame. At certain times of the day the sun may glare into the camera, which would impair its ability to render the scene with good enough definition, and may even damage its sensor.
  • Include just the area you want to capture in the frame; don’t include any irrelevant areas as this simply wastes the resolution of your camera.

Here is an example of a bad camera angle for motion detection:

Bad Camera Angle

Bad Camera Angle For Motion Detection

Continue reading →