Two research strands, each identifying an area of markedly increasing importance in the current development of pattern analysis technology, underlie the review covered by this paper, and are drawn together to offer both a task-oriented and a fundamentally generic perspective on the discipline of pattern recognition. The first of these is the concept of decision fusion for high-performance pattern recognition, where (often very diverse) classification technologies, each providing complementary sources of information about class membership, can be integrated to provide more accurate, robust and reliable classification decisions. The second is the rapid expansion in technology for the automated analysis of (especially) handwritten data for OCR applications including document and form processing, pen-based computing, forensic analysis, biometrics and security, and many other areas, especially those which seek to provide online or offline processing of data which is available in a human-oriented medium. Classifier combination/multiple expert processing has a long history, but the sheer volume and diversity of possible strategies now available suggest that it is timely to consider a structured review of the field. Handwritten character processing provides an ideal context for such a review, both allowing engagement with a problem area which lends itself ideally to the performance enhancements offered by multi-classifier configurations, but also allowing a clearer focus to what otherwise, because of the unlimited application horizons, would be a task of unmanageable proportions. Hence, this paper explicitly reviews the field of multiple classifier decision combination strategies for character recognition, from some of its early roots to the present day. In order to give structure and a sense of direction to the review, a new taxonomy for categorising approaches is defined and explored, and this both imposes a discipline on the presentation of the material available and helps to clarify the mechanisms by which multi-classifier configurations deliver performance enhancements. The review incorporates a discussion both of processing structures themselves and a range of important related topics which are essential to maximise an understanding of the potential of such structures. Most importantly, the paper illustrates explicitly how the principles underlying the application of multi-classifier approaches to character recognition can easily generalise to a wide variety of different task domains.

Multiple classifier decision combination strategies for character recognition: A review

Apparatus for determining a distance between two handwritten strings in a database. A processor extracts global features from each string. The processor divides the string into strokes, and identifies a plurality of bounding boxes. Each box contains a different stroke. The processor extracts global features from the suing, including: (1) a number of points; (2) a maximum angle between a first point in the string and a corner of the tallest bounding box; (3) a number of positive inversions; and (4) a number of negative inversions. The apparatus calculates the distance between the strings based on all of the numbers of points, maximum angles, numbers of positive inversions and numbers of negative inversions. A fixed query tree index may be formed. The tree has leaves and internal nodes belonging to multiple levels. A different key is associated with each level. Each key is a handwritten string. Each string is associated with one of the leaves, such that each child of each internal node in any of the levels between the one leaf and the root node is a root of a respective subtree. Each string associated with any leaf in the subtree which includes the one leaf is equally distant from the key associated with the one level, based on the global features. The tree is queried to search for a subset of the strings, such that each string in the subset is within a threshold distance of an input string, according to the distance function.

Method and apparatus for similarity matching of handwritten data objects

Apparatus and processes are described for the automatic recognition of alphanumeric images. A set of cuts are made to the image which include incorrect segmentations. The resulting "cells" comprising in their totality the created segments of the image are then analyzed to determine which cells are legal neighbors and which are not. All cells which are legal neighbors are then presented as connected nodes. A pruning of nodes which are related to certain predetermined image cuts is effected. Each set of remaining connected nodes is then presented to a recognizer which identifies the image and assigns a specified probability to the output. Many cells which are not legal neighbors are thereby not presented to the recognizer, thus saving substantially on computations per recognized image.

Graphical system for automated segmentation and recognition for image recognition systems

A method for indexing a plurality of handwritten objects is provided. A B-tree data structure of order m is generated, where m is an integer. The B-tree has a plurality of nodes divided into a plurality of levels ordinally numbered 0 th through n th . Each node in the 0 th level is a leaf. Each node in the 1 th level has at least m/2 leaves as children. Each one of the handwritten objects is assigned to a respective leaf. A respectively different hidden Markov model (HMM) is associated with each respective child of each of the nodes in the 1 th to n th levels. Each one of the nodes in the 1 th to n th levels contains the respective HMM associated with the child of the one node. Each HMM in each one of the nodes in the 1 th level is trained to accept the handwritten object of the respective leaf that is a child of the one node. Each HMM associated with any of the nodes in the 2 th through n th levels is trained to accept all of the handwritten objects in the leaves of a subtree that has as a root the respective child with which the HMM is associated.

Method and apparatus for indexing a plurality of handwritten objects

The speed and accuracy of a computer implemented handwriting recognition system is enhanced by several innovations, including integrated segmentation and context processing. The recognition processing occurs while the user is providing ink data. The system quickly reaches the recognition result once all of the input is received. More than one result may be returned by the system.

Handwriting recognition system simultaneously considering shape and context information

A hierarchical system for character recognition with hidden-Markov model knowledge sources that solves both the context-sensitive problem and the character-instantiation problem is presented. The algorithms and the structure of the system are described, and its operation is discussed. The system achieves 97 to 99% accuracy using a two-level architecture and has been implemented using a systolic array, thus permitting real-time (1 ms per character) multifont and multisize printed character recognition as well as handwriting recognition. >

J.A. Vlontzos

Papers

A hierarchical system for character recognition