The aim of this paper is to define a new family of regular languages, the Locally Threshold Testable Languages in Strict Sense (LTTSS). This family includes the well known family of locally testable languages in strict sense (LTSS) and is included in the family of locally threshold testable languages (LTT). Membership of a word to a LTTSS language can be decided by means of local scanning, using a sliding window of a fixed length k, although in this case, we have to care that the number of occurrences of certain segments of length < k in the words is not greater than a level of restriction, less than a threshold r.
As LTTSS languages may be of interest in disciplines such as P~tternRecognition and specially in Speech Recognition, an inference algorithm that identifies the family of (k, r)-TTSS languages from positive data in the limit is proposed. Finally we also report some results aiming to reflect the evolution of the behavior of this algorithm for different values of k and r when it is used in a handwritten digits recognition task.
A holistic classification system for off-line recognition of legal amounts in checks is described in this paper. The binary images obtained from the cursive words are processed following the human visual system, employing a Hough transform method to extract perceptual features. Images are finally coded into a bidimensional feature map representation. Multilayer perpeptrons are used to classify these feature maps into one of the 32 classes belonging to the CENPARMI database. To select a final classification system, ROC graphs are used to fix the best threshold values of the classifiers to obtain the best tradeoff between accuracy and misclassification.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.