An object detector based on convolutional neural network (CNN) has been widely used in the field of computer vision because of its simplicity and efficiency. The average accuracy of CNN model detection results in the object detector is greatly affected by the loss function. The precision of the localization algorithm in the loss function is the main factor affecting the result. Based on the complete intersection over union (CIoU) loss function, an improved penalty function is proposed to improve the localization accuracy. Specifically, the algorithm more comprehensively considers matching bounding boxes between prediction with ground truth, using the proportional relationship of the aspect ratio from both bounding boxes. Under the same aspect ratio of the two bounding boxes, the influence factors of the prediction box on localization accuracy were considered. In this way, the function of the penalty function is strengthened, and localization accuracy of the network model improved. This loss function is called Improved CIoU (ICIoU). Experiments on the Udacity, PASCAL VOC, and MS COCO datasets have demonstrated the effectiveness of ICIoU in improving localization accuracy of network models by using the one-stage object detector YOLOv4. Compared with CIoU, the proposed ICIoU improved average precision (AP) by 0.57% and AP75 by 0.12% on Udacity, AP by 0.26% and AP75 by 1.28% on PASCAL VOC, and AP by 0.06% and AP75 by 0.65% on MS COCO.

No abstract

In general, most of koreans usually have used Hangul and MS-word as a Word Processor. We studied about the functional efficiency by comparing and analyzing Text Mode, Equation Mode, Table Mode, Frame Editor & OLE Mode, Chart Mode and the other Modes to evaluate the common utility used Hangul 97 and MS-word 97 as a Word Processor. As a result, File size, Drag & Drop, Compression and storage etc. have been recognized characters remarkably in each case. We studied about Hangul 97, MS-word 97 as a user for evaluation criteria items . As a matter of fact, we used Tool Book and Director. So, we designed, developed and materialized S/W "garbage generation, separation, and collection" and then we also compared and examined those Text Mode, Sound Mode, Summation Algorithm Mode and Moving image Mode. In consequence, a user's point of view, characters of Tool Book and Director are showed remarkably through the findings of the questionnaire.

This study was aimed at developing an automatic exam system as an intelligent and high-quality higher math examination solution for Department of Computer Engineering at Pai Chai University (PCU) based on S/W Engineering. As a research and development (R&D) project, the study used the rational unified process (RUP) method for software development. RUP describes how to effectively use commercial and reliable methods to develop and deploy software system. It is a heavyweight process; therefore, it is particularly suitable for large teams to develop large projects. This paper describes the RUP process of this R&D project what we named it as Automatic Higher Mathematically Exam System (AHMES). AHMES provides a new way to automatically generate exam questions. In the study, our team and the requesting department collected the requirements and selected methods and tools. Then, our team designed and implemented the framework of the system, completed the development of some main functions, tested and summarized the system, and planned the future work.