• 한국음향학회지
• /
• 제19권1호
• /
• pp.4-10
• /
• 2000

본 논문은 자동 음소 분할기의 음소 경계 오류를 보상하기 위한 후처리(Postprocessing)에 관한 연구이다. 자동 분절 경계의 오류 범위를 줄일 수 있는 후처리기를 제안하고, 자동 분절 결과를 직접 합성 단위로 사용할 수 있는 대량의 합성용 운율데이터 베이스 구축에 유용함을 기술한다. 제안된 후처리기는 수작업으로 보정된 데이터의 특징벡터를 다층 신경회로망(MLP: Multi-layer perceptron)을 통해 학습을 한 후, 자동 분절 결과와 MLP 기반 후처리를 이용하여 새로운 음소 경계를 추출한다. 우선, 특징벡터 set은 음성학적 지식이 최대한 반영되도록 선정되었다. 그리고, 경계를 추출하기 위해서 비선형 패턴분리에 탁월한 성능을 보이는 MLP를 이용한다. MLP는 매우 다양하게 나타나는 음소 경계간 음성학적 특징을 단시간 내에 적용할 수 있기 때문이다. 마지막으로, 음운환경별로 특징 벡터가 적용되는 제안된 후처리 알고리즘을 이용하여 자동 분절의 경계 오류에 대한 보상이 이루어진다. 문장 단위로 발화된 합성용 데이터베이스에서 후처리기로 보정된 분절 결과는 음성 언어 번역 시스템의 분할율보다 약 19.9%의 향상된 성능을 보였으며, 절대오류 (｜Hand label position-Auto label position｜)는 약 28.6% 감소되었다.

• 한국멀티미디어학회논문지
• /
• 제9권5호
• /
• pp.529-541
• /
• 2006

본 논문에서는 객체기반 비디오 부호화 또는 멀티미디어 편집을 위한 반지동 비디오 객체 분할방식을 제안한다. 반자동 객체분할은 사용자 지원에 의한 분할 방식으로, 비디오 시퀀스의 초기 프레임에서 사용자가 관심객체의 경계를 표시하고 이후의 영상 프레임의 객체를 배경으로부터 연속적으로 분리해 낸다. 제안된 방식은 부분적으로 사용자 조력에 의한 프레임내 분할과 완전 자동에 의한 프레임간 분할 처리과정으로 구성되는데, 영상 전체에 대해 연산을 수행하는 기존 방식과는 달리 객체 경계가 존재하는 영상영역 부분에서만 연산을 수행한다. 프레임내 분할은 사용자가 관심객체의 경계를 지정하고, 이 경계 주위 화소들의 유사성을 이용한 후처리에 의해 정확한 초기 객체를 구한다. 프레임간 분할에서는 이전 프레임에서 추출한 객체의 경계 정보에 근거하여 시간적 유사성을 구한 후 경계와 영역 추적에 의해 연속적으로 동영상 객체를 추출한다. 실험결과로부터 제안된 방식은 비디오 편집, 객체기반 비디오 압축 및 인덱싱 등의 멀미디어 응용에 사용 가능할 정도로 안정되고 정확한 객체추출을 수행함을 확인하였다. 이 결과를 바탕으로 다수의 편리한 기능을 포함한 비디오 편집시스템을 개발하였다.

• International Journal of Computer Science & Network Security
• /
• 제23권12호
• /
• pp.151-160
• /
• 2023

Breast cancer is the most dangerous and deadly form of cancer. Initial detection of breast cancer can significantly improve treatment effectiveness. The second most common cancer among Indian women in rural areas. Early detection of symptoms and signs is the most important technique to effectively treat breast cancer, as it enhances the odds of receiving an earlier, more specialist care. As a result, it has the possible to significantly improve survival odds by delaying or entirely eliminating cancer. Mammography is a high-resolution radiography technique that is an important factor in avoiding and diagnosing cancer at an early stage. Automatic segmentation of the breast part using Mammography pictures can help reduce the area available for cancer search while also saving time and effort compared to manual segmentation. Autoencoder-like convolutional and deconvolutional neural networks (CN-DCNN) were utilised in previous studies to automatically segment the breast area in Mammography pictures. We present Automatic SegmenAN, a unique end-to-end adversarial neural network for the job of medical image segmentation, in this paper. Because image segmentation necessitates extensive, pixel-level labelling, a standard GAN's discriminator's single scalar real/fake output may be inefficient in providing steady and appropriate gradient feedback to the networks. Instead of utilising a fully convolutional neural network as the segmentor, we suggested a new adversarial critic network with a multi-scale L1 loss function to force the critic and segmentor to learn both global and local attributes that collect long- and short-range spatial relations among pixels. We demonstrate that an Automatic SegmenAN perspective is more up to date and reliable for segmentation tasks than the state-of-the-art U-net segmentation technique.

• 한국전자거래학회지
• /
• 제7권1호
• /
• pp.187-206
• /
• 2002

The purpose of this study is to review the process for the trade settlement system between all parties involved in the commercial transaction under SURF System. SURF application is the latest value added service from bolero.net and a fully automated documentary settlement system. It extends bolero.net's capability to enable trade transactions by providing a delivery versus payment system. It has been designed as a standard, shared, infrastructure component for handling trade settlement within an electronic environment to keep step with the operation of electronic Bolero bill of Lading. It exploits the services provided by the Bolero Core Messaging Platform - secure, guaranteed transactions based on boleroXML standards to provide document compliance services with optional bank guarantees. So, SURF system can automatically check all commonly used trade documents such as commercial invoices, bills of lading and certificates of weight and analysis, etc and supports a full range of settlement options including Open Account, Documentary Collections, Documentary Credit. 58 have key features as follows, ① automatic document compliance checking, ② integrated with the Title Registry, ③ supports various forms of vender financing, ④ governed by a set of legally binding rules, ⑤ fully prepared for Straight-Through-Processing, ⑥ Enabled compliance with UCP 500.

• 한국정보시스템학회지:정보시스템연구
• /
• 제6권1호
• /
• pp.223-256
• /
• 1997

Since the business scales of retailing companies become to be very large and the number of items dealt increases explosively, automation of inventory management becomes one of the most important issues to solve in retailing industry. In order to accomplish this automation of inventory management, there must be a great need to a method which can perform real-time decision making on inventory control in an automatic fashion, while communicating with inventory information systems like POS system and automatic warehousing system. But even in this circumstance, there are also many obstructions to such automation like varying demands, limited capacity of warehouse and exhibition room, need for strategic consideration on inventory control, etc., in a real sense. Due to these reasons, it seems very difficult that most large-scaled retailing companies get fully automated inventory management system. To overcome those difficulties and reflect them into inventory control, we propose a automated inventory control methodology for retailing industry based on neural network and policy model. Especially, policy model is devised to deal with dynamic varying demands and using this model, strategic goals on inventory can be considered into inventory control mechanism. Our proposed approach is implemented in workstation and its performance is also empirically verified also against to real case of one of the major retailing firm in Korea.

• Journal of Mechanical Science and Technology
• /
• 제17권5호
• /
• pp.654-667
• /
• 2003

This paper presents a consequence of the systematic approach to identify the aerodynamic parameters of an unmanned aerial vehicle (UAV) equipped with the automatic flight control system. A 3-2-1-1 excitation is applied for the longitudinal mode while a multi-step input is applied for lateral/directional excitation. Optimal time step for excitation is sought to provide the broad input bandwidth. A fully automated programmed flight test method provides high-quality flight data for system identification using the flight control computer with longitudinal and lateral/directional autopilots, which enable the separation of each motion during the flight test. The accuracy of the longitudinal system identification is improved by an additional use of the closed-loop flight test data. A constrained optimization scheme is applied to estimate the aerodynamic coefficients that best describe the time response of the vehicle. An appropriate weighting function is introduced to balance the flight modes. As a result, concurrent system models are obtained for a wide envelope of both longitudinal and lateral/directional flight maneuvers while maintaining the physical meanings of each parameter.

• 대한의용생체공학회:의공학회지
• /
• 제15권4호
• /
• pp.447-454
• /
• 1994

기하학적인 모델을 이용한 좌심실에 대한 단축 2D 심초음파도에서의 심내벽 윤관석의 완전 자동검출방법을 제안하였다. 이 방법은 3단계의 과정에 걸쳐 수행된다. 첫번째 단계는 부표본화된 영상에 초기 중심 추정 알고리듬을 적용하여 초기 중심을 추정한다. 두번째 단계에서는 원 영상에 중심추성 알고리듬을 적용하여 정확한 중심과 타원추정에 필요한 정보를 구한다음, 심내벽에 근사하는 타원모델을 추정한다. 세번째 단계는 추정한 타원모델에 근거한 코스트 함수로서 심내벽 윤관석을 검출한다. 제안된 방법을 조작자의 수동적인 개입없이도 효과적인 결과를 보여준다.

• 한국축산시설환경학회지
• /
• 제6권2호
• /
• pp.113-119
• /
• 2000

A Robotic milking cluster system with the manipulator for an automatic milking system was designed and built for farmer to work easily and comfortably during milking processing. The cluster system was composed of screws, cams and links for power transmission, DC motors, the Quick Basic one-chip microprocessor, the vision system for image processing, and tea-cups. Software, written in Visual C+ and Quick Basic, combined the function of image capture, image processing, milking cluster control, and control into one control. The unit was made to transfer from four fixed points to four teats with four teat-cups. Performance tests of the cluster unit, the fully integrated system, were conducted to attach and detach the teat-cup on the teat of a artificial cow. The transfer programming provided for a teat-cup milking loop during the system starts and comes back the original fixed point at the manipulator of it for milking. It transferred the teat-cup with a success rate of more than 70%. The average time it took ot perform the milking loop was about 20 seconds.

• 한국측량학회지
• /
• 제25권6_1호
• /
• pp.575-583
• /
• 2007

Surface segmentation aims to represent the terrain as a set of bounded and analytically defined surface patches. Many previous segmentation methods have been developed to extract planar patches from LIDAR data for building extraction. However, most of them were not fully satisfactory for more general applications in terms of the degree of automation and the quality of the segmentation results. This is mainly caused from the limited information derived from LIDAR data. The purpose of this study is thus to develop an automatic method to perform surface segmentation by combining not only LIDAR data but also images. A region-based method is proposed to generate a set of planar patches by grouping LIDAR points. The grouping criteria are based on both the coordinates of the points and the corresponding intensity values computed from the images. This method has been applied to urban data and the segmentation results are compared with the reference data acquired by manual segmentation. 76% of the test area is correctly segmented. Under-segmentation is rarely founded but over-segmentation still exists. If the over-segmentation is mitigated by merging adjacent patches with similar properties as a post-process, the proposed segmentation method can be effectively utilized for a reliable intermediate process toward automatic extraction of 3D model of the real world.

• 농업과학연구
• /
• 제41권3호
• /
• pp.251-258
• /
• 2014

In order to set up the basic environmental control systems which the new concept greenhouses have to equip, greenhouse characteristics, environmental management and control systems in domestic glasshouses and plastic houses were investigated and analyzed comparatively. Survey results on the width, length, eaves height, and the number of spans etc. showed that glasshouses were bigger than plastic houses significantly. New concept greenhouses claim to be plastic houses, but it will be reasonable to follow the specifications of the glasshouse. Specifications to be applied to new concept greenhouses were proposed as follows; hot water heating systems, aluminum screens as the thermal curtain, evaporative cooling systems, roof vents on the ridge, circulation fans, $CO_2$ enrichment, hydroponic systems, and automatic irrigation control systems. Environmental measurement systems for the indoor and outdoor temperature, humidity, light, wind speed and indoor $CO_2$ concentration have to be fully equipped. The automatic control system has to be as a complex environmental control system, not a single item control system. Also, for stable dissemination, domestically producing complete greenhouse control system should be made as soon as possible.