• Journal of IKEEE
• /
• v.25 no.3
• /
• pp.493-500
• /
• 2021

In this paper, we propose the development of deep learning structure to improve quality of polygonal containers. The deep learning structure consists of a convolution layer, a bottleneck layer, a fully connect layer, and a softmax layer. The convolution layer is a layer that obtains a feature image by performing a convolution 3x3 operation on the input image or the feature image of the previous layer with several feature filters. The bottleneck layer selects only the optimal features among the features on the feature image extracted through the convolution layer, reduces the channel to a convolution 1x1 ReLU, and performs a convolution 3x3 ReLU. The global average pooling operation performed after going through the bottleneck layer reduces the size of the feature image by selecting only the optimal features among the features of the feature image extracted through the convolution layer. The fully connect layer outputs the output data through 6 fully connect layers. The softmax layer multiplies and multiplies the value between the value of the input layer node and the target node to be calculated, and converts it into a value between 0 and 1 through an activation function. After the learning is completed, the recognition process classifies non-circular glass bottles by performing image acquisition using a camera, measuring position detection, and non-circular glass bottle classification using deep learning as in the learning process. In order to evaluate the performance of the deep learning structure to improve quality of polygonal containers, as a result of an experiment at an authorized testing institute, it was calculated to be at the same level as the world's highest level with 99% good/defective discrimination accuracy. Inspection time averaged 1.7 seconds, which was calculated within the operating time standards of production processes using non-circular machine vision systems. Therefore, the effectiveness of the performance of the deep learning structure to improve quality of polygonal containers proposed in this paper was proven.

• Korean Journal of Heritage: History & Science
• /
• v.54 no.2
• /
• pp.120-135
• /
• 2021

Changdeokgung Rear Garden is an important place to show the essence of the garden culture of the Joseon Dynasty. In the garden landscape experience, the restoration of the road completes the system of connecting the main spaces. Therefore, the restoration of the road requires accurate understanding of its creation, extinction, and maintenance. The purpose of this study was to detail the changes in the path that occurred in the Changdeokgung Palace Rear Garden from the late Joseon Dynasty to the modern and contemporary period by analyzing literature and drawing materials. For a time-series analysis, "Donggwoldo" and "Donggwoldohyeong" produced in the Joseon Dynasty, along with "Changdeokgung Plan Drawing" produced in modern and contemporary times, and aerial photographs were used. Drawings and photographs of different coordinate systems were transformed into one coordinate system in the geographic information system ArcGIS to compare changes in the movements of different periods. The results of the study are as follows. First, a total of 37 sections have been used since Japanese colonial era, of which 13 have been maintained, 14 have disappeared, and 10 have been newly established. Among the extinction sections, the road north of Neungheojeong Pavilion is considered to be an urgent place to connect the space to the garden and restore it to enjoy the scenery. In the new section, it seems necessary to establish a new alternative road or shorten the section for the connecting section between Daebodan and Okryucheon. Second, it was revealed that the biggest and most frequent changes to the road system in the garden were Japanese colonial era and renovations in the 1970s. It is worth noting the changes in the road since the 1970s, rather than Japanese colonial era, where it was difficult to manage the gardens independently. The access road to Okryucheon remained in its original shape until the 1990s, but it was renovated to its current shape due to misperception of the original shape. A project is needed to find out the cause of the change in this period and restore the damaged original shape. The biggest achievement of this study is that it revealed the changes in the garden path of Changdeokgung Palace in modern and contemporary times. The biggest achievement of this study is that it revealed the changes in the road of Changdeokgung Palace Rear Gardens in modern and contemporary times. However, there is a limitation that it has not been able to clearly present the location and shape that should be restored because it has not found data on landscaping plans or maintenance. In order to restore the road using the data revealed in this study, it seems necessary to consider realistic problems such as current space utilization, viewing system, disaster prevention and maintenance.

• The korean journal of orthodontics
• /
• v.36 no.5
• /
• pp.339-348
• /
• 2006

Objective: With development of the skeletal anchorage system, orthodontic mini-implant (OMI) assisted on masse sliding retraction has become part of general orthodontic treatment. But compared to the emphasis on successful anchorage preparation, the control of anterior teeth axis has not been emphasized enough. Methods: A 3-D finite element Base model of maxillary dental arch and a Lingual tipping model with lingually inclined anterior teeth were constructed. To evaluate factors influencing the axis of anterior teeth when OMI was used as anchorage, models were simulated with 2 mm or 5 mm retraction hooks and/or by the addition of 4 mm of compensating curve (CC) on the main archwire. The stress distribution on the roots and a 25000 times enlarged axis graph were evaluated. Results: Intrusive component of retraction force directed postero-superiorly from the 2 mm height hook did not reduce the lingual tipping of anterior teeth. When hook height was increased to 5 mm, lateral incisor showed crown-labial and root-lingual torque and uncontrolled tipping of the canine was increased.4 mm of CC added to the main archwire also induced crown-labial and root-lingual torque of the lateral incisor but uncontrolled tipping of the canine was decreased. Lingual tipping model showed very similar results compared with the Base model. Conclusion: The results of this study showed that height of the hook and compensating curve on the main archwire can influence the axis of anterior teeth. These data can be used as guidelines for clinical application.