• Proceedings of the Korea Information Processing Society Conference
• /
• 2019.10a
• /
• pp.952-953
• /
• 2019

최근 딥러닝으로 진행되는 광학 문자 인식 분야는 대부분 단어 단위로 인식하는 것으로 글자 단위의 영역을 검출하는 데에는 적합하지 못하다. 본 연구는 각 글자의 영역을 검출하기 위해 기존의 딥러닝을 이용한 광학 문자 인식 절차인 단어 분리 과정과 단어 인식 과정을 유지하면서 어텐션 중심을 이용하여 각 글자의 영역을 보다 정확하게 검출하는 것을 목표로 한다. 제안하는 모델은 CRAFT 와 Attention Network 를 사용한 OCR 과정을 확장한 모델로 각 단어 문자열 결과물에 각 글자의 영역을 추가로 나타내게 되며 각 글자와 라벨 간의 IOU 평균은 0.671 로 나타났다.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.9 no.4
• /
• pp.444-450
• /
• 1999

This paper deals with several methods: the clustering method that uses k-means algorithm to abstract the area of characters on the image document and the distance function that suits for the HIS coordinate system to cluster the image. For the prepossessing step to recognize this, or the method of characters segmentate, the algorithm to abstract a discrete character is also proposed, using the linking picture element. This algorithm provides the feature that separates any character such as the touching or overlapped character. The methods of projecting and tracking the edge have so far been used to segment them. However, with the new method proposed here, the picture element extracts a discrete character with only one-time projection after abstracting the character string. it is possible to pull out it. dividing the area into the character and the rest (non-character). This has great significance in terms of processing color documents, not the simple binary image, and already received verification that it is more advanced than the previous document processing system.

• The KIPS Transactions:PartD
• /
• v.17D no.1
• /
• pp.59-66
• /
• 2010

Traditional OCR engines are designed to the scanned documents in calibrated environment. Three dimensional perspective distortion and smooth distortion in images are critical problems caused by un-calibrated devices, e.g. image from smart phones. To meet the growing demand of character recognition of texts embedded in the photos acquired from the non-calibrated hand-held devices, we address the problem in three categorical aspects: rotational invariant method of text region extraction, scale invariant method of text line segmentation, and three dimensional perspective mapping. With the integration of the methods, we developed an OCR for camera-captured images.

• Journal of Korea Society of Industrial Information Systems
• /
• v.8 no.2
• /
• pp.14-20
• /
• 2003

This paper presents a text extraction method of camera based document image. It is more difficult to recognize camera based document image in comparison with scanner based image because of segmentation problem due to variable lighting condition and versatile fonts. Both document binarization and character extraction are important processes to recognize camera based document image. After converting color image into grey level image, gray level normalization is used to extract character region independent of lighting condition and background image. Local adaptive binarization method is then used to extract character from the background after the removal of noise. In this character extraction step, the information of the horizontal and vertical projection and the connected components is used to extract character line, word region and character region. To evaluate the proposed method, we have experimented with documents mixed Hangul, English, symbols and digits of the ETRI database. An encouraging binarization and character extraction results have been obtained.

• Journal of KIISE:Software and Applications
• /
• v.31 no.1
• /
• pp.8-19
• /
• 2004

In this paper, we propose an efficient method for locating destination address block on both of machine-Printed and handwritten Korean mail piece images. The proposed method extracts connected components from the binary mail piece image, generates text lines by merging them, and then groups the text fines into nine clusters. The destination address block is determined by selecting some clusters. Considering the geometric characteristics of address information on Korean mail piece, we split a mail piece image into nine areas with an equal size. The nine clusters are initialized with the center coordinate of each area. A modified Manhattan distance function is used to compute the distance between text lines and clusters. We modified the distance function on which the aspect ratio of mail piece could be reflected. The experiment done with live Korean mail piece images has demonstrated the superiority of the Proposed method. The success rate for 1, 988 testing images was about 93.56%.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.10 no.3
• /
• pp.430-436
• /
• 2006

In this paper, a QT_rev protocol is proposed for identifying all the tags within the identification range. The proposed QT_rev protocol revises the QT protocol, which has a memoryless property. In the QT_rev protocol, the tag will send the remaining bits of their identification codes when the query string matches the first bits of their identification codes. After the reader receives all the responses of the tags, it knows which bit is collided. If the collision occurs in the last bit, the reader can identify two tags simultaneously without further query. According to the simulation results, the QT_rev protocol outperforms the QT protocol in terms of the number of queries and the number of response bits.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• v.9 no.2
• /
• pp.491-494
• /
• 2005

In this paper, a QT_rev protocol is proposed for identifying all the tags within the identification range. The proposed QT_rev protocol revises the QT protocol, which has a memoryless property. In the QT_rev protocol, the tag will send the remaining bits of their identification codes when the query string matches the first bits of their identification codes. After the reader receives all the responses of the tags, it knows which bit is collided. If the collision occurs in the last bit, the reader can identify two tags simultaneously without further query.

• The KIPS Transactions:PartB
• /
• v.10B no.4
• /
• pp.429-442
• /
• 2003

In this paper, we describe an address image recognition system for automatic processing of standard- size letter mail. The inputs to the system are gray-level mail piece images and the outputs are delivery point codes with which a delivery sequence of carrier can be generated. The system includes five main modules; destination address block location, text line separation, character segmentation, character recognition and finally address interpretation. The destination address block is extracted on the basis of experimental knowledge and the line separation and character segmentation is done through the analysis of connected components and vortical runs. For recognizing characters, we developed MLP-based recognizers and dynamical programming technique for interpretation. Since each module has been implemented in an independent way, the system has a benefit that the optimization of each module is relatively easy. We have done the experiment with live mail piece images directly sampled from mail sorting machine in Yuseong post office. The experimental results prove the feasibility of our system.

• Journal of KIISE:Software and Applications
• /
• v.31 no.9
• /
• pp.1121-1130
• /
• 2004

This paper proposes a method for line and word segmentation for machine-printed text blocks. To separate a text region into the unit of lines, it analyses the horizontal projection profile and performs a recursive projection profile cut method. In the word segmentation, between-word gaps are identified by a hierarchical clustering method after finding gaps in the text line by using a connected component analysis. In addition, a special symbol detection technique is applied to find two types of special symbols tying between words using their morphologic features. An experiment with 84 text regions from English and Korean documents shows that the proposed method achieves 99.92% accuracy of word segmentation, while a commercial OCR software named Armi 6.0 Pro$^{TM}$ has 97.58% accuracy.y.

• The KIPS Transactions:PartB
• /
• v.12B no.4 s.100
• /
• pp.369-378
• /
• 2005

Document image is segmented and classified into text, picture, or table by a document layout analysis, and the words in table regions are significant for keyword spotting because they are more meaningful than the words in other regions. This paper proposes a method to extract words from table regions in document images. As word extraction from table regions is practically regarded extracting words from cell regions composing the table, it is necessary to extract the cell correctly. In the cell extraction module, table frame is extracted first by analyzing connected components, and then the intersection points are extracted from the table frame. We modify the false intersections using the correlation between the neighboring intersections, and extract the cells using the information of intersections. Text regions in the individual cells are located by using the connected components information that was obtained during the cell extraction module, and they are segmented into text lines by using projection profiles. Finally we divide the segmented lines into words using gap clustering and special symbol detection. The experiment performed on In table images that are extracted from Korean documents, and shows $99.16\%$ accuracy of word extraction.