Grain Limits
RSI Search uses features of different sizes to locate similar features in run-time images. RSI Search uses large features to perform the initial coarse scan of the image, and it uses small features to precisely locate features.
For example, when RSI Search searches for a trained model of a diskette, it uses the large features from the diskette (such as the overall shape of the diskette and the outline of the label) to quickly locate the diskette, then it uses the smaller features (such as the letters on the label) to determine the precise location of the model. The figure below shows how RSI Search uses the different feature sizes.
Large features used for coarse location and small features for fine location
The features that RSI Search detects in an image are controlled by the granularity that it uses to analyze the image. To detect only the large features in an image, RSI Search uses a larger granularity setting. To detect the small features in an image, RSI Search uses a smaller granularity.
Granularity is expressed as the radius of interest, in pixels, within which features are detected. The figure above illustrates two important characteristics of model granularity.
- Large features such as the outline of the diskette are detected at both small and large granularity settings.
- Smaller features are present or absent from the image depending on the granularity setting.
In some cases, however, a feature might be present at a fine granularity and at a coarse granularity, but not at an intermediate granularity.
RSI Search uses a range of model granularities when it trains a model from an image; RSI Search automatically determines the optimum granularity settings when it trains a model. The smallest granularity used to detect features in the training image or shape description is called the fine granularity limit. The largest granularity used to detect features is called the coarse granularity limit.
You can override the grain limits that RSI Search determines, but in most cases the automatically computed limits provide good performance and accuracy.