• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.16 no.3
• /
• pp.117-122
• /
• 2023

Gaze estimation, which automatically recognizes where a user is currently staring, and object detection based on estimated gaze point, can be a more accurate and efficient way to understand human visual behavior. in this paper, we propose a method to detect the objects within the region of interest around the gaze point. Specifically, after estimating the 3D gaze point, a region of interest based on the estimated gaze point is created to ensure that object detection occurs only within the region of interest. In our experiments, we compared the performance of general object detection, and the proposed object detection based on region of interest, and found that the processing time per frame was 1.4ms and 1.1ms, respectively, indicating that the proposed method was faster in terms of processing speed.

• Journal of Korean Tunnelling and Underground Space Association
• /
• v.26 no.2
• /
• pp.129-152
• /
• 2024

In the application of deep learning object detection via CCTV in tunnels, a large number of false positive detections occur due to the poor environmental conditions of tunnels, such as low illumination and severe perspective effect. This problem directly impacts the reliability of the tunnel CCTV-based accident detection system reliant on object detection performance. Hence, it is necessary to reduce the number of false positive detections while also enhancing the number of true positive detections. Based on a deep learning object detection model, this paper proposes a false positive data training method that not only reduces false positives but also improves true positive detection performance through retraining of false positive data. This paper's false positive data training method is based on the following steps: initial training of a training dataset - inference of a validation dataset - correction of false positive data and dataset composition - addition to the training dataset and retraining. In this paper, experiments were conducted to verify the performance of this method. First, the optimal hyperparameters of the deep learning object detection model to be applied in this experiment were determined through previous experiments. Then, in this experiment, training image format was determined, and experiments were conducted sequentially to check the long-term performance improvement through retraining of repeated false detection datasets. As a result, in the first experiment, it was found that the inclusion of the background in the inferred image was more advantageous for object detection performance than the removal of the background excluding the object. In the second experiment, it was found that retraining by accumulating false positives from each level of retraining was more advantageous than retraining independently for each level of retraining in terms of continuous improvement of object detection performance. After retraining the false positive data with the method determined in the two experiments, the car object class showed excellent inference performance with an AP value of 0.95 or higher after the first retraining, and by the fifth retraining, the inference performance was improved by about 1.06 times compared to the initial inference. And the person object class continued to improve its inference performance as retraining progressed, and by the 18th retraining, it showed that it could self-improve its inference performance by more than 2.3 times compared to the initial inference.

• Journal of the Korean School Mathematics Society
• /
• v.10 no.2
• /
• pp.263-288
• /
• 2007

The purpose of this study is to search the effective teaching-learning program by considering how affect on formation of the process-object perspective of linear function using Excel. In this study we analyzed function units in textbook and examined how Excel affect on the formation of the process-object perspective of linear function. Teaching experiment was based on qualitative case study and performed for five classes with five 8th graders. Data were gathered through observations, audio-taped interviews, video recording of the students 'work, students' worksheets, and detailed field notes. Findings indicate that exploration learning environment using Excel could supplement paper-and-pencil environment. We found that intuitive, dynamic, explorative, feedback skills via Excel can play the role of scaffolding supporting formation of process perspective object perspective of linear function.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.8
• /
• pp.2688-2703
• /
• 2022

Research on the advanced detection of harmful objects in airport cargo for passenger safety against terrorism has increased recently. However, because associated studies are primarily focused on the detection of relatively large objects, research on the detection of small objects is lacking, and the detection performance for small objects has remained considerably low. Here, we verified the limitations of existing research on object detection and developed a new model called the Small Hazardous Object detection enhanced and reconstructed Model based on the You Only Look Once version 5 (YOLOv5) algorithm to overcome these limitations. We also examined the performance of the proposed model through different experiments based on YOLOv5, a recently launched object detection model. The detection performance of our model was found to be enhanced by 0.3 in terms of the mean average precision (mAP) index and 1.1 in terms of mAP (.5:.95) with respect to the YOLOv5 model. The proposed model is especially useful for the detection of small objects of different types in overlapping environments where objects of different sizes are densely packed. The contributions of the study are reconstructed layers for the Small Hazardous Object detection enhanced and reconstructed Model based on YOLOv5 and the non-requirement of data preprocessing for immediate industrial application without any performance degradation.

• 한국문화재보존과학회:학술대회논문집
• /
• 2002.11a
• /
• pp.85-89
• /
• 2002

• Journal of the Korea Society of Computer and Information
• /
• v.21 no.5
• /
• pp.79-89
• /
• 2016

In this paper, we proposed the object classification method using genetic and dynamic random forest consisting of optimal combination of unit tree. The random forest can ensure good generalization performance in combination of large amount of trees by assigning the randomization to the training samples and feature selection, etc. allocated to the decision tree as an ensemble classification model which combines with the unit decision tree based on the bagging. However, the random forest is composed of unit trees randomly, so it can show the excellent classification performance only when the sufficient amounts of trees are combined. There is no quantitative measurement method for the number of trees, and there is no choice but to repeat random tree structure continuously. The proposed algorithm is composed of random forest with a combination of optimal tree while maintaining the generalization performance of random forest. To achieve this, the problem of improving the classification performance was assigned to the optimization problem which found the optimal tree combination. For this end, the genetic algorithm methodology was applied. As a result of experiment, we had found out that the proposed algorithm could improve about 3~5% of classification performance in specific cases like common database and self infrared database compare with the existing random forest. In addition, we had shown that the optimal tree combination was decided at 55~60% level from the maximum trees.

• Proceedings of the Korean Society of Medical Physics Conference
• /
• 2002.09a
• /
• pp.113-115
• /
• 2002

The Monte Carlo simulation method is a numerical solution to a problem that models objects interacting with other objects or their environment based upon simple object-object or object-environment relationships. In spite of its great accuracy, It was turned away because of long calculation time to simulate a model. But, it is used to simulate a linear accelerator frequently with the advance of computer technology. To simulate linear accelerator in Monte Carlo simulations, there are many parameters needed to input to Monte Carlo code. These data can be supported by a linear accelerator manufacturer. Although the model of a linear accelerator is the same, a different characteristic property can be found. Thus, we performed a commissioning process of 6MV photon beam in Varian 2300C/D model with BEAM/EGS4 Monte Carlo code. The head geometry data were put into BEAM/EGS4 data. The mean energy and energy spread of the electron beam incident on the target were varied to match Monte Carlo simulations to measurements. TLDs (thermoluminescent dosimeter) and radiochromic films were employed to measure the absorbed dose in a water phantom. Beam profile was obtained in 40cm${\times}$40cm field size and Depth dose was in 10cm${\times}$10cm. At first, we compared the depth dose between measurements and Monte Carlo simulations varying the mean energy of an incident electron beam. Then, we compared the beam profile with adjusting the beam radius of the incident electron beam in Monte Carlo simulation. The results were found that the optimal mean energy was 6MV and beam radius of 0.1mm was well matched to measurements.

• Science of Emotion and Sensibility
• /
• v.11 no.4
• /
• pp.501-510
• /
• 2008

The purpose of this study was to investigate whether the orientation of the head position across different categories affect reaction time and accuracy of object recognition. Fifty four right handed undergraduate students were participated in the experiment. Participants performed the word-picture matching tasks, which were different in terms of head direction of object (i.e., Left-headed or Right-headed) and object category (i.e., natural : animal or artificial : tool). Participants were asked to decide whether each picture matched the word which was followed by the picture. For accuracy, no statistically significant difference was found for both animal and tool pictures due to the ceiling effect. Interaction effect of category and orientation were statistically significant, whereas only the main effect of category was significant. In the animal condition, faster reaction times were observed for left to right than right to left presentation, while no statistical significant difference was found in the tool condition. The orientation of the object's canonical representation was different across different categories. The faster RT for the animal condition implies that the canonical representation for animal is left-headed. This could be due to the orientation of the face.

• Korean Institute of Interior Design Journal
• /
• v.25 no.2
• /
• pp.21-29
• /
• 2016

This study has estimated the traits of conscious eye moving with the objects of the halls of subway stations. For that estimation, the observation data from eye-tracking were matched with the experiment images, while an independent program was produced and utilized for the analysis of the eye moving in the selected sections, which could provide the ground for clarifying the traits of space-users' eye moving. The outcomes can be defines as the followings. First, The application of the independently produced program provides the method for coding the great amount of observation data, which cut down a lot of analysis time for finding out the traits of conscious eye moving. Accordingly, the inclusion of eye's intentionality in the method for extracting the characteristics of eye moving enabled the features of entrance and exit of particular objects with the course of observing time to be organized. Second, The examination of eye moving at each area surrounding the object factors showed that [out]${\rightarrow}$[in], which the line of sight is from the surround area to the objects, characteristically moved from the left-top (Area I) of the selected object to the object while [in]${\rightarrow}$[out], which is from the inside of the object to the outside, also moved to the left-top (Area I). Overall, there were much eye moving from the tops of right and left (Area I, II) to the object, but the eye moving to the outside was found to move to the left-top (Area I), the right-middle (Area IV) and the right-top (Area II). Third, In order to find if there was any intense eye-moving toward a particular factor, the dominant standards were presented for analysis, which showed that there was much eye-moving from the tops (Area I, II) to the sections of 1 and 2. While the eye-moving of [in] was [I $I{\rightarrow}A$](23.0%), [$I{\rightarrow}B$](16.1%) and [$II{\rightarrow}B$](13.8%), that of [out] was [$A{\rightarrow}I$](14.8%), [$B{\rightarrow}I$](13.6%), [$A{\rightarrow}II$](11.4%), [$B{\rightarrow}IV$](11.4%) and [$B{\rightarrow}II$](10.2%). Though the eye-moving toward objects took place in specific directions (areas), that (out) from the objects to the outside was found to be dispersed widely to different areas.

• Conservation Science in Museum
• /
• v.3
• /
• pp.65-69
• /
• 2001

Scientific analysis are carried on gilded ornamental shoes, which have been excavated from tomb No 1. and No 4. in Beopcheon-ri, Wonju dated from Baekje Period. This object is a very important because it provides valuable information on the development of metal-making techniques of that period. Thus, this article illustrates the investigation conducted to reveal how the object was created using what materials and techniques. Instead of the spikes-attached to the bottom plate of the object-that couldn't be sampled, a sample of a decorative rivet showing a similar structure to the spikes-attached to the other parts of the shoes-was prepared for a cross-section examination. Using radiography and microscopes, it was found that the head of a spike was placed and then punched to the gilded plate, so it can be held to the shoes. Under the SEM-EDS, the cross-section of the rivet shows that the gilding layer was applied before the attachment of the rivets. It also shows that the gilding layer is distributed unevenly and there are empty spaces indicating amalgam gilding was employed. This was confirmed as Mercury was detected on the SEM-EDS analysis of the object. The examination of the microstructure of the plate using the SEM-EDS revealed that the object is made of a single copper alloy plate with recrystallized twining and non-metallic white inclusions, which found to be lead in this case.