• 한국지능시스템학회논문지
• /
• 제17권4호
• /
• pp.557-562
• /
• 2007

This paper extends the work of Maji et al. (2001) to present the concept of interval-valued fuzzy soft sets and to present an algorithm for finding where the degree of membership are represented by interval values in [0, 1]. The proposed method is more flexible than the one presented in Maji et at. (2001) due to the fact that it allows the degrees of membership of object for parameters to be represented by interval-values rather than crisp real values between zero and one.

• 대한소아치과학회지
• /
• 제29권3호
• /
• pp.450-454
• /
• 2002

연구개 및 구인두부의 열창은 아동들에게 상대적으로 흔한 편이다. 연구개 열창은 신생아부터 노년층까지 모든 연령에 걸쳐 보고된 바 있지만, 어린 아동들 특히 걸음마를 막 시작한 아동들에게서 많이 보고되었다. 이러한 아동들의 경우 입안에 물체를 물고 돌아다니다가 넘어지거나 떨어져서 구인두부 열창을 입게되는 경우가 많은 편이고, 주로 4세 정도에 호발하며, 여아보다는 남아에서 더 이러한 외상이 자주 나타나는 것으로 보고되었다. 주로 손상을 입는 연구개 및 구인두 부위는 좌측 편도 상단부위이며, 관통상없이 선상이나 표면상의 손상으로 주로 발생하고, 손상 형태는 U자 형태 혹은 V자 형태로 나타난다. 이러한 외상에 있어 관통상이나, 크게 조직편이 이개된 경우를 제외하고는 봉합술은 필요하지 않으며, 오히려 이러한 봉합술이 외상부위의 치유를 지연시킬 수 있다. 본 창상의 치유에는 3주정도 소요되며 상흔은 거의 남지 않는 것으로 보고되었다. 그러나 연구개의 창상이 경동맥의 손상을 야기하여 신경학적인 합병증이 발생했다는 보고가 있으므로, 2-3일간은 외상을 입은 환아에 대한 주기적인 관찰이 필요하며, 합병증 발생시 구강외과의 및 이비인후과의 또는 신경외과의와의 협의진료가 필요하다. 본 증례는 2세 및 3세 남환아로 입안에 이물질을 물고 넘어져 연구개의 열창을 입은 경우로 비외과적인 처치 및 신경학적인 평가 후 완치되어 이에 보고하는 바이다.

• 한국컴퓨터그래픽스학회논문지
• /
• 제7권1호
• /
• pp.1-9
• /
• 2001

소프트 물체 가시화 방법의 개선을 위해, 크기가 적응 분할되는 셀들로 공간을 분할 하는 방법과 자료 구조를 제안한다. 소프트 물체의 가시화를 위해서는 필드 함수 값을 공간상의 모든 점들에서 구할 수 있어야 한다. 본 논문에서는 주어진 한 점에 영향을 미치는 소프트 물체들을 인터벌 트리 (interval tree)라는 자료구조를 검색함으로써 계산할 수 있다. 인터벌트리는 물체들의 제한 용적 (bounding volume)에 기반하며, 점이나 선분들과 같은 단순한 뼈대 (skeleton)로부터 생성된 소프트 물체를 나타내고 있다. 각 구성 컴포넌트 (componet)의 제한 용적은 국지 필드 함수의 반경, 한계 값 (threshold value), 이웃한 컴포넌트들과의 관계 등으로부터 계산된다. 제안된 방법은 소프트 물체의 모델링이나 렌더링 등의 많은 응용에 사용될 수 있으며, 특히 대화형 모델링 과정에 유용하다.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제11권2호
• /
• pp.71-76
• /
• 2011

The problem of decision making under imprecise environments are widely spread in real life decision situations. We present a method of object recognition from imprecise multi observer data, which extends the work of Roy and Maji [J Compu. Appl. Math. 203(2007) 412-418] to generalized intuitionistic fuzzy soft set theory. The method involves the construction of a comparison table from a generalized intuitionistic fuzzy soft set in a parametric sense for decision making.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2020년도 춘계학술발표대회
• /
• pp.468-471
• /
• 2020

본 논문은 ESRGAN(Enhanced Super Resolution GAN)과 Semantic Soft Segmentation을 이용한 객체 분할의 성능 개선에 관한 것이다. 본 논문의 연구진이 이미 제안한 Mask R-CNN과 Semantic Soft Segmentation을 이용한 객체 분할 방법은 전반적으로 객체 분할 성능이 양호한 반면, 객체의 크기가 상대적으로 작으면 분할 성능이 저조해지는 문제점이 있었다. 본 논문은 이러한 문제점을 해결하기 위한 것으로, Mask R-CNN을 통해 검출된 객체의 크기가 일정 기준치 이하인 경우, ESRGAN을 통해 초해상화를 수행한 후, Semantic Soft Segmentation을 수행함으로써 소형 객체의 분할 성능을 개선함에 그 목적이 있다. 제안된 방법에 따르면, 기존의 방볍에 비해 크기가 작은 객체의 분할 특성을 좀 더 효과적으로 개선할 수 있음을 확인할 수 있었다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제9권7호
• /
• pp.2633-2648
• /
• 2015

The problem of visual words' synonymy and ambiguity always exist in the conventional bag of visual words (BoVW) model based object category methods. Besides, the noisy visual words, so-called "visual stop-words" will degrade the semantic resolution of visual dictionary. In view of this, a novel bag of visual words method based on PLSA and chi-square model for object category is proposed. Firstly, Probabilistic Latent Semantic Analysis (PLSA) is used to analyze the semantic co-occurrence probability of visual words, infer the latent semantic topics in images, and get the latent topic distributions induced by the words. Secondly, the KL divergence is adopt to measure the semantic distance between visual words, which can get semantically related homoionym. Then, adaptive soft-assignment strategy is combined to realize the soft mapping between SIFT features and some homoionym. Finally, the chi-square model is introduced to eliminate the "visual stop-words" and reconstruct the visual vocabulary histograms. Moreover, SVM (Support Vector Machine) is applied to accomplish object classification. Experimental results indicated that the synonymy and ambiguity problems of visual words can be overcome effectively. The distinguish ability of visual semantic resolution as well as the object classification performance are substantially boosted compared with the traditional methods.

• 로봇학회논문지
• /
• 제6권3호
• /
• pp.220-229
• /
• 2011

This paper focuses on a development of an anthropomorphic robot hand. Human hand is able to dexterously grasp and manipulate various objects with not accurate and sufficient, but inaccurate and scarce information of target objects. In order to realize the ability of human hand, we develop a robot hand and introduce a control scheme for stable grasping by using only kinematic information. The developed anthropomorphic robot hand, KITECH Hand, has one thumb and three fingers. Each of them has 4 DOF and a soft hemispherical finger tip for flexible opposition and rolling on object surfaces. In addition to a thumb and finger, it has a palm module composed the non-slip pad to prevent slip phenomena between the object and palm. The introduced control scheme is a quitely simple based on the principle of virtual work, which consists of transposed Jacobian, joint angular position, and velocity obtained by joint angle measurements. During interaction between the robot hand and an object, the developed robot hand shows compliant grasping motions by the back-drivable characteristics of equipped actuator modules. To validate the feasibility of the developed robot hand and introduced control scheme, collective experiments are carried out with the developed robot hand, KITECH Hand.

• 한국멀티미디어학회논문지
• /
• 제24권3호
• /
• pp.416-422
• /
• 2021

Due to the large differences in human activity within classes, the large similarity between classes, and the problems of visual angle and occlusion, it is difficult to extract features manually, and the detection rate of human behavior is low. In order to better solve these problems, an improved Faster R-CNN-based detection algorithm is proposed in this paper. It achieves multi-object recognition and localization through a second-order detection network, and replaces the original feature extraction module with Dense-Net, which can fuse multi-level feature information, increase network depth and avoid disappearance of network gradients. Meanwhile, the proposal merging strategy is improved with Soft-NMS, where an attenuation function is designed to replace the conventional NMS algorithm, thereby avoiding missed detection of adjacent or overlapping objects, and enhancing the network detection accuracy under multiple objects. During the experiment, the improved Faster R-CNN method in this article has 84.7% target detection result, which is improved compared to other methods, which proves that the target recognition method has significant advantages and potential.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2003년도 ICCAS
• /
• pp.2682-2686
• /
• 2003

Recent paradigm in technology shifts from object-based technology to environment-based technology. Issue here is interaction among human, machines, and environment. This requires new interpretations for the space among them. Holistic interactions based on “Mom (embodiment)” suggest a good starting point for this endeavor. The past, present, future of technology are presented in terms of technology’s fundamental virtue: “humanizing technology” or “technology serving human.” Interactive technology initiative (ITI) is an interdisciplinary research group to search for the proper technology and the proper way of implementing technology: “interactive technology” or “soft engineering.” Some experimental activities conducted by ITI are presented in this organized session, “Interactive Technology.” Metatechnology, soft engineering, “Mom (embodiment),” holistic interactions, tangible space, and ubiquitous computing are key concepts in interactive technology.

• Journal of Magnetics
• /
• 제22권1호
• /
• pp.29-33
• /
• 2017

We report a micromagnetic numerical study on different quasi-crystal formations of magnetic vortices in a rich variety of dynamic transient states in soft magnetic nano-disks. Only the application of spin-polarized dc currents to a single magnetic vortex leads to the formation of topological-soliton quasi-crystals composed of different configurations of skyrmions with positive and negative half-integer numbers (magnetic vortices and antivortices). Such topological object formations in soft magnets, not only in the absence of Dzyaloshinskii-Moriya interaction but also without magnetocrystalline anisotropy, are discussed in terms of two different topological charges, the winding number and the skyrmion number. This work offers an insight into the dynamic topological-spin-texture quasi-crystal formations in soft magnets.