To train neural networks, it is necessary to collect and submit to AxxonSoft video recordings and images from your actual cameras taken in the same resolution and under the same conditions as in your future application.
For example, if your neural network is intended to analyze outdoor video feeds, your footage must contain all range of weather conditions (sun, rain, snow, fog, etc.) in different times of day (daytime, twilight, night).
General requirements for collected data:
if it is required to train the neural network in different conditions of time of day, lighting, angle, object types or weather, then the video material should be collected in equal shares for each condition, that is, it should be balanced.
Example. It is necessary to detect a person in the surveillance area at night and during the day. Data collected correctly:
Data collected incorrectly:
|
Extra requirements for video footage for each neural analytics tool are listed in the following table:
Tool | Requirements |
Neural Filter | At least 1000 frames containing different objects of interest in the given scene conditions, and the same number of frames containing no objects (noise frames) |
Neural Tracker | 3 to 5 minutes of video containing objects of interest in the given scene conditions. The more the number and variability of the situations in the scene, the better |
Posture detection tools | 10 seconds of video of a scene with no persons. At least 100 different persons in the given scene conditions. Attention! Different conditions mean, among others, different postures of an individual in scene (tilting, different limbs patterns, etc.) |
Personal protective equipment detection tools | A list of all reference equipment with examples should be collected from the object and agreed with the analytics manufacturer (see Example of providing a list of valid equipment at the facility). Several video recordings 3-5 minutes each with personnel in the given scene conditions. Personnel should move and change posture in the collected video recordings, as well as remove and put on equipment at intervals of 30 seconds. Since the Personal protective equipment detection tools are designed for artificial constant lighting, video recordings in other lighting conditions are not required |
| At least 1000 frames containing different objects of interest in the given scene conditions, and the same number of frames containing no objects (noise frames) | |
| Object presence detection tool | At least 1000 frames containing different objects of interest in the given scene conditions, and the same number of frames containing no objects (noise frames) |
Food recognition * | Images of at least 80% of the actual menu items should be provided. Each menu item requires 20 to 40 images shot in different conditions. |
If the above requirements for the collection of data transmitted for training the neural network model are met, and if the neural network is operated in the conditions that are as similar as possible to the conditions in which the material for its training was collected, then the overall accuracy** of neural network analytics is guaranteed from 90% to 97% and the percentage of false positives is 5-7%. For general networks***, an overall accuracy of 80-95% and a false positive rate of 5-20% are guaranteed.
* Will be available in future versions of Axxon One. ** Accuracy is indicated for a neural network model, which was trained under operating conditions. *** A general network is a network that was not trained under operating conditions. |
The requirements may be changed or added to at any time.