Recognition of architectural and electrical symbols by COSFIRE filters with inhibition

The automatic recognition of symbols can be used to automatically convert scanned drawings into digital representations compatible with computer aided design software. We propose a novel approach to automatically recognize architectural and electrical symbols.The proposed method extends the existing trainable COSFIRE approachby adding an inhibition mechanism that is inspired by shape-selectiveTEO neurons in visual cortex. A COSFIRE filter with inhibition takesas input excitatory and inhibitory responses from line and edge detectors. The type (excitatory or inhibitory) and the spatial arrangementof low level features are determined in an automatic configuration stepthat analyzes two types of prototype pattern called positive and negative.Excitatory features are extracted from a positive pattern and inhibitoryfeatures are extracted from one or more negative patterns. In our experiments we use four subsets of images with different noise levels fromthe Graphics Recognition data set (GREC 2011) and demonstrate thatthe inhibition mechanism that we introduce improves the effectivenessof recognition substantially