Automatic Bankcheck Processing

The following sections are included:

• CONTENTS

Bankcheck processing represents an important challenge for the scientific community working in the field of document analysis and handwriting recognition. Difficulties in solving the problem derive mainly from the nature of bankchecks, which is extremely complex intrinsically…

A new bankcheck processing system is presented in this paper. A full exploitation of the contextual knowledge, together with a multi-expert approach, have been used both to analyze the complex shape of handwritten text and to design the system.

Several processing modules have been integrated in the system. Some of the most relevant are those for data acquisition, preprocessing, machine-printed numeral recognition, layout analysis, courtesy amount recognition, legal amount recognition, amount validation, and signature verification. Some combination techniques have also been used in the system.

Reuse and maintenance of the system were two of the main goals of the designing process and the Khoros software tool was used for this purpose.

We developed a check reading system, termed INTERCHEQUE, which recognizes both the legal (LAR) and the courtesy amount (CAR) on bank checks. The version presented here is designed for the recognition of French, omni-bank, omni-scriptor, handwritten bank checks, and meets industrial requirements, such as high processing speed, robustness, and extremely low error rates.

We give an overview of our recognition system and discuss some of the pattern recognition techniques used. We also describe an installation which processes of the order of 70,000 checks per day. Results on a data base of about 170,000 checks show a recognition rate of about 75% for an error rate of the order of 1/10,000 checks.

This paper describes a prototype for Brazilian bankcheck recognition. The description is divided into three topics: bankcheck information extraction, digit amount recognition and signature verification. In bankcheck information extraction, our algorithms provide signature and digit amount images free of background patterns and bankcheck printed information. In digit amount recognition, we dealt with the digit amount segmentation and implementation of a complete numeral character recognition system involving image processing, feature extraction and neural classification. In signature verification, we designed and implemented a static signature verification system suitable for banking and commercial applications. Our signature verification algorithm is capable of detecting both simple, random and skilled forgeries. The proposed automatic bankcheck recognition prototype was intensively tested by real bankcheck data as well as simulated data providing the following performance results: for skilled forgeries, 4.7% equal error rate; for random forgeries, zero Type I error and 7.3% Type II error; for bankcheck numerals, 92.7% correct recognition rate.

In this paper, a complete fault-tolerant check recognition system is proposed which has no check substitution error under the secret code verification. The fault-tolerant recognition method proposed in this paper creates all possible candidates for verification under the limited fault-tolerant rate, and with three classifiers of high isolated digit recognition rate, the system can always find out the correct recognition results of checks if there exist the correct labels of all the digits. Since the three classifiers are designed independently by different methods and they extract different features of handwritten digits, they can compensate each other when confusing digits are met. The segmentation stage combines the three most popular strategies, and gives out a way for segmenting unconstrained handwritten numeral strings on Chinese checks.

This paper presents a complete numeral amount recognition module which is integrated in an automatic system aimed at reading all types of French checks. This module is combined with an automatic reading system of literal amounts. This complete working system, called LIREChèques, is developed by MATRA MS&I and is now in advanced test at SERINTEL, a pilot site.

Two aspects of the numeral amount recognition system are particularly emphasized: the numeral recognition stage itself and the syntactic analysis stage. The numeral recognition module relies on a combination of two individual classifiers, the first one is based on concavity measurements, the second one on both statistical and structural features. The syntactic analysis, called syntactic/contextual analysis, is combined with contextual information to take into account the segmentation behaviour and the presence of literal entities in the numeral amount.

We demonstrate that very good performances can be obtained on digits such as those extracted from numeral amounts since a substitution rate of 0.06% while still preserving a recognition rate of near 87% can be achieved. As for the syntactic/contextual analysis stage, results obtained on a test set (containing checks from more than 40 different banks and 15% of typed checks, thus being a good representation of the real tests realized on site) show clearly that introduction of contextual information in association with syntactic analysis allows to process much more numeral amounts than a simple syntactic analysis and increases perceptibility of the recognition rate.

We describe an implemented system which reads amounts on French checks images. This system is made of two modules which recognize independently the courtesy and the legal amount. The first one is based on a segmentation-by-recognition approach while the second uses hidden Markov models of words. The outputs of these two modules are combined into a system decision. A high reliability is achieved through this combination since the errors made by the recognition modules should be mostly uncorrelated.

Check amount recognition is one of the most promising commercial applications of handwriting recognition. This paper is devoted to the description of the check reading system developed to recognize amounts on American personal checks. Special attention is paid to a reliable procedure developed to reject doubtful answers. For this purpose the legal (worded) amount on a personal check is recognized along with the courtesy (digit) amount. For both courtesy and legal amount fields, a brief description of all recognition stages beginning with field extraction and ending with the recognition itself are presented. We also present the explanation of problems existing at each stage and their possible solutions. The numeral recognizer used to read the amounts written in figures is described. This recognizer is based on the procedure of matching input subgraphs to graphs of symbol prototypes. Main principles of the handwriting recognizer used to read amounts written in words are explained. The recognizer is based on the idea of describing the handwriting with the most stable handwriting elements. The concept of the optimal confidence level of the recognition answer is introduced. It is shown that the conditional probability of the answer correctness is an optimal confidence level function. The algorithms of the optimal confidence level estimation for some special cases are described. The sophisticated algorithm of cross validation between legal and courtesy amount recognition results based on the optimal confidence level approach is proposed. Experimental results on real checks are presented. The recognition rate at 1% error rate is 67%. The recognition rate without reject is 85%. Significant improvement is achieved due to legal amount processing in spite of a relatively low recognition rate for this field.

A bankcheck reading system using cross validation of both the legal and the courtesy amounts is presented in this paper. Some of the challenges posed by the task are

(i) segmentation of the legal amount into words,

(ii) location of boundaries between dollars and cents amounts, and

(iii) high accuracy in terms of recognition performance.

Word segmentation in the legal amount is a serious issue because of the nature of the data and patrons' writing habits which tend to clump words together. We have developed a word segmentation algorithm based on the character segmentation results to address this issue. The list of possible amounts generated by the word segmentation hypotheses is used as lexicon for the courtesy amount recognition. The order of magnitude of the amount is estimated during legal amount recognition. We treat the courtesy amount as a numeral string and apply the same word recognition scheme as used for the legal amount.

Our approach to check recognition differs from traditional methods in two significant aspects: First, our emphasis on both the legal and the courtesy amounts is balanced. We use an accurate word recognizer which performs equally well on alpha words and digit strings. Second, our combination strategy is serial rather than the commonly used parallel method. Experimental results show that 43.8% of check images are correctly read with an error rate of 0%.

A novel approach to extract data from check images is proposed based on the determination of baselines of checks, a priori information about the positions of data on checks, and a layout-driven item extraction method. Several techniques and algorithms have been developed in this approach including check image preprocessing, the extraction and identification of baselines, the extraction of the strokes of handwritten legal amounts, courtesy amounts and date, and the separation of strokes connected to baselines. A complete working system has been developed. The results of both testing experiments and on-line applications show that this approach is effective and the proposed techniques and algorithms perform well.

In this paper we propose a new method to extract user entered components from a personal bankcheck. The proposed technique utilizes a pair of personal bankcheck images: (i) blank check image referred to as the reference image and (ii) a filled-in check image referred to as the sample image. This pair of check images (reference image and sample image) differ only at specific regions of the check reserved for insertion of handwritten (machine printed) components. A sophisticated technique of inter-image subtractions can extract the user entered components. Also the background patterns of reference image and sample images are not identical because of statistical variations in printing and scanning conditions. The authors propose a morphological inter-image subtraction that consists of a GLS-fusion procedure and a logical subtraction procedure in gray-level space (GLS) (x, y, f(x, y)) to extract user entered components from the filled bankcheck. The user entered components are extracted by this method in a very robust manner and the extracted components are suitable for posterior word/string recognition step. This method with morphological subtraction can also be used for other forms.

Neural networks are now widely and successfully used in the recognition of handwritten numerals. Despite their wide use in recognition, neural networks have not seen widespread use in segmentation. Segmentation can be extremely difficult in the presence of connected numerals, fragmented numerals, and background noise, and its failure is a principal cause of rejected and incorrectly read documents. Therefore, strategies leading to the successful application of neural technologies to segmentation are likely to yield important performance benefits. In this paper we identify problems that have impeded the use of neural networks in segmentation and describe an evolutionary approach to applying neural networks in segmentation. Our approach, based upon the use of monotonic fuzzy valued decision functions computed by feed-forward neural networks, has been successfully employed in a production system.

This paper proposes symbolic and neural classifiers to read unconstrained handwritten worded amounts in bankchecks. Features are extracted from the binary image of the worded amount. Depending on the features extracted, some words are recognized entirely symbolically, some words entirely neurally, and the remaining both symbolically and neurally. Results of experiments at word level and check level are provided.

Off-line cursive script recognition has got increasing attention during the last three decades since it is of interest in several areas such as banking and postal service. An off-line cursive handwritten word recognition system is described in this paper and is used for legal amount interpretation in personal checks. The proposed recognition system uses a set of geometric and topologic features to characterize each word. By considering the spatial distribution of these features in a word image, the proposed system maps each word into two strings of finite symbols. A local associative indexing scheme is then used on these strings to organize a vocabulary. When presented with an unknown word, the system uses the same indexing scheme to retrieve a set of candidate words likely to match the input word. A verification process is then carried out to find the best match among the candidate set. The performance of the proposed system has been tested with a legal amount image database from real bankchecks. The results obtained indicate that the proposed system is able to recognize legal amounts with great accuracy.

In this paper we present a system for the recognition of handwritten words on literal check amounts which advantageously combine HMMs and Markov random fields (MRFs). It operates at pixel level, in a holistic manner, on height normalized word images which are viewed as random field realizations. The HMM analyzes the image along the horizontal writing direction, in a specific state observation probability given by the column product of causal MRF-like pixel conditional probabilities. Aspects concerning definition, training and recognition via this type of model are developed throughout the paper. We report a 90.08% average word recognition rate on 2378 words and a 79.52% amount rate on 579 amounts of the SRTP* French postal check database (7031 words, 1779 amounts, different scriptors).

The aim of this study is to show that the optimal order of Markov Model of cursive words can be rigorously stated in order to fit the structural properties of the observed data using Akaike information criterion. The method has been tested on French Postal check amounts up to order 4. An original structural representation of cursive words based on graphemes is used. The conditional probability to have a word model given an observed sequence of graphemes is computed independently of the length of the sequence. The recognition results obtained confirm the optimal order found using Akaike criterion.

This paper presents an Off-line Signature Verification System for identifying random forgeries aimed at banking application. The cognitive information learning process in the proposed system is inspired by some characteristics of human learning. Four features distinguish the proposed system from those proposed thus far. First, the verification task is accomplished without a priori knowledge of the class of random forgeries. Second, no explicit modeling or making geometrical measurements are used to represent the signature. Third, the decision of the system is made throughout the use of two-stage verification process by which a global and/or local analysis are performed on the unknown signature. The global analysis is concerned with the overall shape of the unknown signature, whereas, the local analysis is concerned with the local features composing the unknown signature. Fourth, these analysis are performed at the boundary or within a predefined search region called the identity grid designed for each writer in the system. The proposed system is evaluated with a data base of 800 signatures.

In this paper a multi-expert signature verification system is presented. The system has been specifically designed for applications in the field of bankcheck processing. For this purpose, it combines three different algorithms for signature verification. A wholistic approach is used in the first algorithm, a component-oriented approach is used in the second and third algorithms. The second algorithm is based on a structure-based procedure, the third algorithm uses a highly-adaptive neural network. The three algorithms are combined in the multi-expert system by a voting strategy.