Horst Bunke

The IAM-database: an English sentence database for offline handwriting recognition

Abstract. In this paper we describe a database that consists of handwritten English sentences. It is based on the Lancaster-Oslo/Bergen (LOB) corpus. This corpus is a collection of texts that comprise about one million word instances. The database includes 1,066 forms produced by approximately 400 different writers. A total of 82,227 word instances out of a vocabulary of 10,841 words occur in the collection. The database consists of full English sentences. It can serve as a basis for a variety of handwriting recognition tasks. However, it is expected that the database would be particularly useful for recognition tasks where linguistic knowledge beyond the lexicon level is used, because this knowledge can be automatically derived from the underlying corpus. The database also includes a few image-processing procedures for extracting the handwritten text from the forms and the segmentation of the text into lines and words.

A graph distance metric based on the maximal common subgraph

Abstract Error-tolerant graph matching is a powerful concept that has various applications in pattern recognition and machine vision. In the present paper, a new distance measure on graphs is proposed. It is based on the maximal common subgraph of two graphs. The new measure is superior to edit distance based measures in that no particular edit operations together with their costs need to be defined. It is formally shown that the new distance measure is a metric. Potential algorithms for the efficient computation of the new measure are discussed.

On a relation between graph edit distance and maximum common subgraph

Abstract In approximate, or error-correcting, graph matching one considers a set of graph edit operations, and defines the edit distance of two graphs g1 and g2 as the shortest (or least cost) sequence of edit operations that transform g1 into g2. A maximum common subgraph of two graphs g1 and g2 is a subgraph of both g1 and g2 such that there is no other subgraph of g1 and g2 with more nodes. Graph edit distance and maximum common subgraph are well known concepts that have various applications in pattern recognition and machine vision. In this paper a particular cost function for graph edit distance is introduced, and it is shown that under this cost function graph edit distance computation is equivalent to the maximum common subgraph problem.

Approximate graph edit distance computation by means of bipartite graph matching

In recent years, the use of graph based object representation has gained popularity. Simultaneously, graph edit distance emerged as a powerful and flexible graph matching paradigm that can be used to address different tasks in pattern recognition, machine learning, and data mining. The key advantages of graph edit distance are its high degree of flexibility, which makes it applicable to any type of graph, and the fact that one can integrate domain specific knowledge about object similarity by means of specific edit cost functions. Its computational complexity, however, is exponential in the number of nodes of the involved graphs. Consequently, exact graph edit distance is feasible for graphs of rather small size only. In the present paper we introduce a novel algorithm which allows us to approximately, or suboptimally, compute edit distance in a substantially faster way. The proposed algorithm considers only local, rather than global, edge structure during the optimization process. In experiments on different datasets we demonstrate a substantial speed-up of our proposed method over two reference systems. Moreover, it is emprically verified that the accuracy of the suboptimal distance remains sufficiently accurate for various pattern recognition applications.

Using a statistical language model to improve the performance of an HMM-based cursive handwriting recognition systems

In this paper, a system for the reading of totally unconstrained handwritten text is presented. The kernel of the system is a hidden Markov model (HMM) for handwriting recognition. The HMM is enhanced by a statistical language model. Thus linguistic knowledge beyond the lexicon level is incorporated in the recognition process. Another novel feature of the system is that the HMM is applied in such a way that the difficult problem of segmenting a line of text into individual words is avoided. A number of experiments with various language models and large vocabularies have been conducted. The language models used in the system were also analytically compared based on their perplexity.

Inexact graph matching for structural pattern recognition

This paper is concerned with the inexact matching of attributed, relational graphs for structural pattern recognition. The matching procedure is based on a state space search utilizing heuristic information. Some experimental results are reported.

A new algorithm for error-tolerant subgraph isomorphism detection

We propose a new algorithm for error-correcting subgraph isomorphism detection from a set of model graphs to an unknown input graph. The algorithm is based on a compact representation of the model graphs. This representation is derived from the set of model graphs in an off-line preprocessing step. The main advantage of the proposed representation is that common subgraphs of different model graphs are represented only once. Therefore, at run time, given an unknown input graph, the computational effort of matching the common subgraphs for each model graph onto the input graph is done only once. Consequently, the new algorithm is only sublinearly dependent on the number of model graphs. Furthermore, the new algorithm can be combined with a future cost estimation method that greatly improves its run-time performance.

Offline recognition of unconstrained handwritten texts using HMMs and statistical language models

This paper presents a system for the offline recognition of large vocabulary unconstrained handwritten texts. The only assumption made about the data is that it is written in English. This allows the application of statistical language models in order to improve the performance of our system. Several experiments have been performed using both single and multiple writer data. Lexica of variable size (from 10,000 to 50,000 words) have been used. The use of language models is shown to improve the accuracy of the system (when the lexicon contains 50,000 words, the error rate is reduced by /spl sim/50 percent for single writer data and by /spl sim/25 percent for multiple writer data). Our approach is described in detail and compared with other methods presented in the literature to deal with the same problem. An experimental setup to correctly deal with unconstrained text recognition is proposed.

IAM Graph Database Repository for Graph Based Pattern Recognition and Machine Learning

In recent years the use of graph based representation has gained popularity in pattern recognition and machine learning. As a matter of fact, object representation by means of graphs has a number of advantages over feature vectors. Therefore, various algorithms for graph based machine learning have been proposed in the literature. However, in contrast with the emerging interest in graph based representation, a lack of standardized graph data sets for benchmarking can be observed. Common practice is that researchers use their own data sets, and this behavior cumbers the objective evaluation of the proposed methods. In order to make the different approaches in graph based machine learning better comparable, the present paper aims at introducing a repository of graph data sets and corresponding benchmarks, covering a wide spectrum of different applications.

Edge Detection in Range Images Based on Scan Line Approximation

In this paper we present a novel edge detection algorithm for range images based on a scan line approximation technique. Compared to the known methods in the literature, our algorithm has a number of advantages. It provides edge strength measures that have a straightforward geometric interpretation and supports a classification of edge points into several subtypes. We give a definition of optimal edge detectors and compare our algorithm to this theoretical model. We have carried out extensive tests using real range images acquired by four range scanners with quite different characteristics. Using a simple contour closure technique, we show that our edge detection method is able to achieve a complete range image segmentation into regions. This edge-based segmentation approach turns out to be superior to many region-based methods with regard to both segmentation quality and computational efficiency. The good results that were achieved demonstrate the practical usefulness of our edge detection algorithm.