Pusan National University

About: Pusan National University is a education organization based out in Busan, South Korea. It is known for research contribution in the topics: Catalysis & Population. The organization has 24124 authors who have published 45054 publications receiving 819356 citations. The organization is also known as: Busan National University & Pusan University.

YOLO-Tomato: A Robust Algorithm for Tomato Detection Based on YOLOv3

Abstract: Automatic fruit detection is a very important benefit of harvesting robots. However, complicated environment conditions, such as illumination variation, branch, and leaf occlusion as well as tomato overlap, have made fruit detection very challenging. In this study, an improved tomato detection model called YOLO-Tomato is proposed for dealing with these problems, based on YOLOv3. A dense architecture is incorporated into YOLOv3 to facilitate the reuse of features and help to learn a more compact and accurate model. Moreover, the model replaces the traditional rectangular bounding box (R-Bbox) with a circular bounding box (C-Bbox) for tomato localization. The new bounding boxes can then match the tomatoes more precisely, and thus improve the Intersection-over-Union (IoU) calculation for the Non-Maximum Suppression (NMS). They also reduce prediction coordinates. An ablation study demonstrated the efficacy of these modifications. The YOLO-Tomato was compared to several state-of-the-art detection methods and it had the best detection performance.

Clinically Suspected Pulmonary Embolism: Utility of Spiral CT

Abstract: PURPOSE: To prospectively determine the utility of contrast material–enhanced spiral computed tomography (CT) in the examination of patients clinically suspected of having pulmonary embolism (PE). MATERIALS AND METHODS: One hundred ten patients clinically suspected of having PE were examined with contrast-enhanced spiral CT and at least one other imaging modality: ventilation-perfusion scintigraphy, Doppler ultrasonography of deep leg veins, or pulmonary angiography. Chart review or telephone contact with the referring clinician was used to evaluate the contribution of spiral CT to the final clinical diagnosis. RESULTS: Spiral CT helped correctly identify 23 of 25 patients with PE (sensitivity, 92%). In 57 (67%) of the 85 patients without PE, spiral CT provided additional information that suggested or confirmed the alternate clinical diagnosis: pneumonia (n = 14), cardiovascular disease (n = 10), pulmonary fibrosis (n = 7), trauma (n = 6), malignancy (n = 5), pleural disease (n = 4), postoperative changes...

Online binary decision decoding using functional near-infrared spectroscopy for the development of brain–computer interface

Abstract: In this paper, a functional near-infrared spectroscopy (fNIRS)-based online binary decision decoding framework is developed. Fourteen healthy subjects are asked to mentally make “yes” or “no” decisions in answers to the given questions. For obtaining “yes” decoding, the subjects are asked to perform a mental task that causes a cognitive load on the prefrontal cortex, while for making “no” decoding, they are asked to relax. Signals from the prefrontal cortex are collected using continuous-wave near-infrared spectroscopy. It is observed and verified, using the linear discriminant analysis (LDA) and the support vector machine (SVM) classifications, that the cortical hemodynamic responses for making a “yes” decision are distinguishable from those for making a “no” decision. Using mean values of the changes in the concentration of hemoglobin as features, binary decisions are classified into two classes, “yes” and “no,” with an average classification accuracy of 74.28 % using LDA and 82.14 % using SVM. These results demonstrate and suggest the feasibility of fNIRS for a brain–computer interface.