Nai-Jen Cheng

On-line chinese character recognition with effective candidate radical and candidate character selections

Abstract A useful property of Chinese characters is that most of them possess a radical with a meaning. Our system is motivated by this characteristic found in Chinese characters which leads us to radical-based candidate selection approaches, performed before similarity measurement. The searching ranges of the radicals and the characters can be basically decided by the number of strokes in the input script. From the recognized possible radicals, the candidate characters are limited to those having such radicals and those having no radical. Since some of the candidate radicals may have very high matching costs, we can reduce the recognition time by discarding those unlikely candidate radicals. Furthermore, to speed up the inspection of radicals, we developed a radical extraction algorithm to narrow down the searching scope of the reference templates. The radical extraction algorithm was further improved by eliminating false extracted radicals. With these mechanisms, the number of candidate radicals screened out is 286 out of 726 radical templates and the number of candidate characters to be detailed matched is 123 out of 5401 Chinese characters. Through these efforts, the recognition rate improves to be 96.35% for the first rank and 98.96% for the first 10 result candidates with the speed of 0.427 seconds on average per character on a PC using Intel 386-33 CPU.

On-line Chinese signature verification with mixture of experts

We propose an on-line Chinese signature verification system based on mixture of experts to further improve the reliability of a signature verification system. As we know, the signature of a certain signer has some personal particular characteristics. We collect these characteristics, and classify them into several feature vectors. In our system, an individual feature vector corresponds to an expert which performs the function of signature verification. According to the particular habit of an individual signer, an individual expert with a particular feature vector plays a different role with different power to a different signer. These verification results should be weighted by a weighted controller. The function of the weighted controller works as the personal habit function. In our proposed system, there are seven verification experts with different feature vectors to make their verification decision. Based on these results, the combination unit with a weighted controller makes the final decision. The weighted controller controls the weight of the decision between the experts and the individual signer. Then, the combination of mixture results of the seven experts is attained, and a signature verification system with a personal-oriented feature decision is approached. Finally, the performance of our system is demonstrated in several experiments which show a high success rate of verification.

On-line Chinese signature verification using voting scheme

We present a system for online Chinese signature verification which is based on the voting result of nine verification experts. First, we briefly describe the individual verification experts. Then, a voting method for expert combination is discussed. Finally, the performance of the combination is demonstrated in several practical experiments which show a high success rate of verification.

Handwritten Chinese signature verification based on attributed string matching of stroke linkage order

We propose an attributed string matching approach based on the writing sequences of an input signature for Chinese signature verification. It is impossible to find features of a signature that are invariant with respect to individual writing style. The aim of our research is to find a particular feature set that will exhibit small intraclass variance. A signer tends to connect consecutive strokes in a constant sequences when signing a signature, especially for Chinese signatures. Therefore, the writing sequences (stroke order) of a signature can be regarded as the personal signature feature. In order to obtain an attributed string that will be used in the string matching similarity calculation, we must split an input signature into several segments from the corners of strokes at first. Since the wavelet transform has been used in the field of edge detection and corner detection for a long time, we can use it as a tool for corner detection to find an optimal segmentation set. Therefore, an attributed string for consecutive segments can be calculated after the signature has been split into several segments suitably by means of a wavelet transform. The elements in the attributed vector include relative angle of adjoining segments, the direction code of the segment, writing duration time of the segment, and the length of the segment. The total relative angle can also be summarized over all of the relative angles of segments to result in another individual feature. This total relative angle can be used to filter the rougher forgery signatures. Our attributed string matching method is also based on the extraction of irreducible characteristic points. The experimental results show a very excellent discrimination capability.

Real-time incoherent optical pattern recognition with photorefractive crystals

Incoherent optical pattern recognition has been demonstrated with a simple setup using photorefractive BaTiO3 crystal. The input patterns to be identified can be transmitted into the system with complete incoherent light. The setup does not required pre-fabricated filters for its operation and thus is a real-time system. The working principles of the systems are the photorefractive beam fanning effect and the degenerate four-wave mixing.

Generation of o-polarized phase-conjugated waves using photogalvanic effect

We have demonstrated a double phase conjugator with orthogonal incident polarizations that generates both extraordinary and ordinary polarized conjugate waves simultaneously in the same photorefractive crystal, with its polarization state preserved. The e-conjugate waves are yielded by photorefractive backscattering in the crystal, and the o-conjugate waves take place due to the coupling of the incident e-waves and o-waves through circular photogalvanic effect.

Phase conjugator with high-o/e-ratio tolerance using the BaTiO3 crystal

Self-pumped photorefractive phase conjugators only work with extra-ordinarily polarized waves (e-waves) with respect to the photorefractive crystal used in the conjugator. This is because photorefractive effect is prominent only with e-wave in crystals. Ordinarily polarized waves (o-waves) do not only give rise to this kind of photorefractive effect but often suppress the e-wave photorefractive effect when both of them are directed into the crystal. In this paper, we propose another photorefractive phase conjugator, which although works with e-waves too, is very insensitive to the suppressing effect of a beam of o-waves occupying the same propagation channel.

Incoherent multifunction parallel optical digital computing device in double self-pumped phase conjugators

A multi-function optical computing system performing the NOR and NAND operations with two photorefractive BaTiO3 crystals is proposed. Instead of using the interference between the input signals and local incoherent erasing, we use incoherent erasing techniques in the construction of this system. The multi-function system works with different polarization states of the input optical waves. The system works with different polarization states of the input optical waves. The system is able to accept temporally incoherent and spatially incoherent signals. Optical experimental result with spatially incoherent optical images demonstrating the incoherent parallel processing capabilities and the multi-function logic operations of the system are shown.

An intelligent Chinese pen-based secure screen phone for new PSTN services

This article presents an intelligent Chinese secure screen phone for new public switching telephone network (PSTN) services. Built upon the most frequently used global telephone network, this new screen phone can provide a lot of additional new services that an old phone does not have. These services include the client equipment for home banking, teleshopping and the deaf-dumb communicator etc.

Matching algorithm for radical-based online Chinese character recognition

In this paper, we propose a matching algorithm for radical-based on-line Chinese character recognition. The major effort of this paper is to demonstrate recognition procedures for subcharacters, such as radicals and residual subcharacters, and nonradical characters. Since a Chinese character could have front radical, rear radical or none of them, the matching algorithm should be able to take care of all these conditions. Furthermore, instead of picking up the front/rear radical strokes from the input character before the matching process is taken, our matching algorithm determines how many strokes the front/rear radical should be during the matching process; it thus enjoys the property of flexibility. After the radical type and the number of strokes of the radical are figured out, the residual subcharacter can be picked up and submitted for matching again. By sequentially recognizing the types of front/rear radicals and the types of residual subcharacters, we can determine what the input characters are.