Jonghak Lee

Robust optical measurement of water vapor in steel-making process

Based on tunable diode laser spectroscopy (TDLS), we develop a simple and efficient humidity monitoring scheme to improve reliability of measurement under randomly changing transmission condition. This measurement scheme can detect the humidity of a monitored system in either high or low scattering condition such as powder coal fired furnace in steel-making process. Distributed feed-back (DFB) diode laser is used to evaluate the relative and absolute humidity by making direct absorption measurement near 938 nm region where 2nu1 + nu3 water vapor transition band exists. This system gives reliable measurement results under either high or low scattering condition.

Vision technique for the recognition of billet characters in the steel plant

This paper describes vision technique about the real time billet characters recognition system in the steel production line. Normally, the billets are mixed at yard so that their identifications are very difficult and very important processing. The character recognition algorithm used in this paper is base on the subspace method by K-L transformation. With this method, we need no special feature extraction steps, which are usually error prone. So the gray character images are directly used as input vectors of the classifier. To train the classifier, we have extracted eigen vectors of each character, which was included in the billet images. We have constructed vision system for the recognition of billet characters using this algorithm and tested this system in the steel production line. The recognition rate of our system in the field test has turned out to be 98.6 % if the corrupted characters are excluded. In the results, we have confirmed that our recognition system has a good performance in the poor environments and ill-conditioned marking system such as steel production plant.

Development of real-time character recognition system for the steel-iron plant

This paper describes about the development of real-time recognition system for the characters, which was marked at the slab and billet material in the steel-iron plant. Normally, the billets or slabs are mixed at yard so that their identifications are hardly tractable. The characters of these material image can be marked by use three kinds of methods such as auto marking machine, template-marking plate and hand written method. For the recognition of each character, we use KLT, PSV, and CCD & PSVM algorithm for the printed font, template-marked font and hand written font respectively. The font style of slab materials is printer letter. The font style of bloom or billet materials is mixed type with printed font, template-marked font and hand written font. For the application of recognition algorithm, we have developed vision system and installed at the slab and billet processing line of the steel-iron plant. Through the test, we have confirmed that our system has good performance and high recognition rate, 97.6% for slab and 98.6% for billet

Reliable optical measurement of water vapor in highly scattering environment

Based on tunable diode laser spectroscopy with direct absorption measurement, we developed a reliable measurement scheme which can precisely detect water vapor concentration independent of the scattering (attenuation) caused by the non-water dust particles inside the observed humidity chamber. The relative water vapor concentration was measured at 2nu(1)+nu(3) water vapor transition band by 938 nm distributed feed-back diode laser at different temperatures using the tunable diode laser spectroscopy technique. These relative water vapor concentrations are converted into the absolute water vapor concentrations and we confirmed that the experimental results at different temperatures are consistently independent of the attenuation. Our measurement system gives the consistent water vapor concentrations without regard to either high or low scattering caused by the non-water dust particles inside the observed humidity chamber.

DEVELOPMENT OF RECOGNITION SYSTEM FOR TEMPLATE-MARKED DIGITS IN BILLETS

Abstract This paper presents new solutions to recognize template-marked digits in billets. Since these digits can be easily blurred, systematic preprocessing methods and recognition algorithms with better generalization properties are required. In this paper, preprocessing methods using the local threshold and skew correction are proposed to extract features with good quality from captured billet images. And multi-class Support Vector Machine(SVM) based algorithms which select the final class among potential classes are applied. Experimental results show that the proposed approach has higher recognition rate than that of the previous method using the K-Nearest Neighborhood(K-NN) technique for template-marked digits in billets.

Surface Inspection System Using the Real-Time Image Processing

Abstract We have developed an innovative surface inspection system for automated quality control for steel productions in POSCO. We had ever installed the various kinds of surface inspection systems. such as a linear CCD and a laser typed surface inspection system at cold rolled strip production lines. But these systems cannot fulfill the sufficient detection and classification rate. and real time processing performance. In order to increase detection and classification rate. we have used the Dark. Bright and Transition field illumination and area type CCD camera and for the real time image processing parallel computing has been used. In this paper we introduce the automatic surface inspection system and real time image processing technique using the object detection, defect detection, classification algorithm and its application to automatic quality control system.