• KIPS Transactions on Software and Data Engineering
• /
• v.1 no.2
• /
• pp.115-126
• /
• 2012

The artificial captions appearing in videos include information that relates to the videos. In order to obtain the information carried by captions, many methods for caption extraction from videos have been studied. Most traditional methods of detecting caption region have used one frame. However video include not only spatial information but also temporal information. So we propose a method of detection caption region using temporal and spatial information. First, we make improved Text-Appearance-Map and detect continuous candidate regions through matching between candidate-regions. Second, we detect disappearing captions using disappearance test in candidate regions. In case of captions disappear, the caption regions are decided by a merging process which use temporal and spatial information. Final, we decide final caption regions through ANNs using edge direction histograms for verification. Our proposed method was experienced on many kinds of captions with a variety of sizes, shapes, positions and the experiment result was evaluated through Recall and Precision.

• 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.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2007.06a
• /
• pp.161-163
• /
• 2007

기존의 텍스트 추출을 위한 이진화 방법은 입력 이미지를 명도 이미지로 변환한 뒤 이진화 하는 방법을 사용하였다. 이러한 방법은 칼라 이미지에서는 극명히 구분되는 색이라 할지라도 명도 이미지로 변환하는 과정에서 같은 밝기를 같게 되는 경우(예를 들어, 배경은 붉은색, 텍스트는 초록색), 텍스트를 추출하는 데 어려움이 있다. 본 논문에서는 이러한 문제를 해결하기 위해 입력 이미지를 R, G, B로 분리하고 각각을 이진화 하여 텍스트를 추출하고 다해상도 웨이블릿(Wavelet) 변환을 이용하여 텍스트의 획 특징을 추출하여 추출된 특징들을 SVM(Support Vector Machine) 분류기로 검증하여 최종 텍스트 영역을 확정한다. 제안한 방법을 적용함으로써 명도 정보만으로는 추출하기 어려웠던 텍스트 영역을 효과적으로 추출하고 텍스트와 구별하기 어려운 영역을 획수준으로 검증할 수 있었다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2021.05a
• /
• pp.418-421
• /
• 2021

모바일 디바이스 화면에 대하여 클릭 가능한 객체를 인식하기 위한 Object detection network architecture 를 제안한다. DSSD 를 Baseline 으로 SE block 이 추가된 Backbone network 와 SSD layer, FPN 구조를 사용한다. 기존의 1:1 비율의 네트워크의 Input resolution 을 모바일 화면과 유사한 1:2 비율로 변경하여 효율적으로 피처를 추출한다. 또한 해당 모델을 학습하기 위한 효율적인 데이터셋을 구축한다. 모바일 화면에서 클릭 가능한 객체를 기준으로 데이터를 수집하여 총 24,937 개의 Annotation data 를 Text, Image, Button, Region 등 8 개의 카테고리로 세분화하였다.

• Journal of Korea Multimedia Society
• /
• v.18 no.7
• /
• pp.843-852
• /
• 2015

Binarization of document images is a critical pre-processing step required for character recognition. Even though various research efforts have been devoted, the quality of binarization results largely depends on the noise amount and condition of images. We propose a new binarization method that combines Maximally Stable External Region(MSER) with down-sampling. Particularly, we propose to apply different threshold values for character regions, which turns out to be effective in reducing noise. Through a set of experiments on test images, we confirmed that the proposed method was superior to existing methods in reducing noise, while the increase of execution time is limited.

• The Journal of the Acoustical Society of Korea
• /
• v.40 no.5
• /
• pp.466-472
• /
• 2021

Speaker Change Detection (SCD) refers to finding the moment when the main speaker changes from one person to the next in a speech conversation. In speaker change detection, difficulties arise due to overlapping speakers, inaccuracy in the information labeling, and data imbalance. To solve these problems, TIMIT corpus widely used in speech recognition have been concatenated artificially to obtain a sufficient amount of training data, and the detection of changing speaker has performed after identifying overlapping speakers. In this paper, we propose an speaker change detection system that considers the speaker overlapping. We evaluated and verified the performance using various approaches. As a result, a detection system similar to the X-Vector structure was proposed to remove the speaker overlapping region, while the Bi-LSTM method was selected to model the speaker change system. The experimental results show a relative performance improvement of 4.6 % and 13.8 % respectively, compared to the baseline system. Additionally, we determined that a robust speaker change detection system can be built by conducting related studies based on the experimental results, taking into consideration text and speaker information.

• Journal of Broadcast Engineering
• /
• v.19 no.5
• /
• pp.578-589
• /
• 2014

Window image is displayed through a monitor screen when we execute the application programs on the computer. This includes webpage, video player and a number of applications. The webpage delivers a variety of information by various types in comparison with other application. Unlike a natural image captured from a camera, the window image like a webpage includes diverse components such as text, logo, icon, subimage and so on. Each component delivers various types of information to users. However, the components with different characteristic need to be divided locally, because text and image are served by various type. In this paper, we divide window images into many sub blocks, and classify each divided region into background, text and subimage. The detected subimages can be applied into 2D-to-3D conversion, image retrieval, image browsing and so forth. There are many subimage classification methods. In this paper, we utilize AdaBoost for verifying that the machine learning-based algorithm can be efficient for subimage detection. In the experiment, we showed that the subimage detection ratio is 93.4 % and false alarm is 13 %.

• Journal of Broadcast Engineering
• /
• v.13 no.4
• /
• pp.544-553
• /
• 2008

With the development of video editing technology, there are growing uses of overlay text inserted into video contents to provide viewers with better visual understanding. Since the content of the scene or the editor's intention can be well represented by using inserted text, it is useful for video information retrieval and indexing. Most of the previous approaches are based on low-level features, such as edge, color, and texture information. However, existing methods experience difficulties in handling texts with various contrasts or inserted in a complex background. In this paper, we propose a novel framework to localize the overlay text in a video scene. Based on our observation that there exist transient colors between inserted text and its adjacent background a transition map is generated. Then candidate regions are extracted by using the transition map and overlay text is finally determined based on the density of state in each candidate. The proposed method is robust to color, size, position, style, and contrast of overlay text. It is also language free. Text region update between frames is also exploited to reduce the processing time. Experiments are performed on diverse videos to confirm the efficiency of the proposed method.

• Proceedings of the KAIS Fall Conference
• /
• 2012.05b
• /
• pp.716-718
• /
• 2012

본 논문에서는 모바일 폰으로 획득한 영상의 텍스트영역 검출을 제안한다. 최근 모바일 폰을 이용한 영상기반 응용 분야의 연구가 활발히 진행되고 있으며, 특히 영상에서 텍스트를 인식하기 위한 전단계로 텍스트 영역 검출은 중요하다. 본 논문은 텍스트 영역 검출을 위해 먼저, 컬러 영상을 입력 받아 그레이 이미지로 변환하여 영상내에 내포된 잡음을 제거하고 열림/닫힘 연산의 특징을 이용해 각 연결요소를 검출하고 검출된 요소들을 레이블링 한다. 레이블링 된 영상은 텍스트가 갖는 특정 조건에 의해 텍스트 영역인지 텍스트 영역이 아닌지를 검출하고 검출된 텍스트 영역은 검증을 통해 최종 텍스트 영역을 검출한다. 제안한 방법은 기존의 택스트 영역 겁출보다 정확도가 향상할 수 있다.

• Proceedings of the KAIS Fall Conference
• /
• 2007.05a
• /
• pp.181-184
• /
• 2007

자연영상에 포함된 텍스트는 많은 중요한 정보를 포함하고 있으므로 자연영상에서 텍스트 정보를 검출하는 연구가 활발히 진행되고 있다. 본 논문에서는 문자 영역의 구조적인 특정을 배열문법으로 정의한 적응적 문자-에지 맵을 제안하여 텍스트 영역을 검출한다. 캐니-에지 검출기로 에지를 추출하고, 생성된 에지 이미지를 레이블링하고 그 영역의 문자구조 특징을 분석하기 위해서 적응적 문자-에지 맵을 분석한다. 적응적 문자-에지 랩의 분포 상태를 분석함으로서 텍스트 후보 영역을 검출하고, 텍스트 영역의 에지 히스토그램 프로파일을 분석함으로서 텍스트 후보 영역에 대한 검증을 수행하여 최종적인 텍스트 영역을 검출한다. 제안한 방법은 다양한 종류의 자연영상을 대상으로 실험하였고, 기울어진 텍스트와 다양한 크기의 텍스트 구성된 자연영상에서 텍스트 영역을 효과적으로 검출하였다.