• Journal of Advanced Research in Ocean Engineering
• /
• v.3 no.3
• /
• pp.136-148
• /
• 2017

The paper presents dual arm ROV manipulation using deep reinforcement learning. The purpose of this underwater manipulator is to investigate and excavate natural resources in ocean, finding lost aircraft blackboxes and for performing other extremely dangerous tasks without endangering humans. This research work emphasizes on a self-learning approach using Deep Reinforcement Learning (DRL). DRL technique allows ROV to learn the policy of performing manipulation task directly, from raw image data. Our proposed architecture maps the visual inputs (images) to control actions (output) and get reward after each action, which allows an agent to learn manipulation skill through trial and error method. We have trained our network in simulation. The raw images and rewards are directly provided by our simple Lua simulator. Our simulator achieve accuracy by considering underwater dynamic environmental conditions. Major goal of this research is to provide a smart self-learning way to achieve manipulation in highly dynamic underwater environment. The results showed that a dual robotic arm trained for a 3DOF movement successfully achieved target reaching task in a 2D space by considering real environmental factor.

• Journal of the Optical Society of Korea
• /
• v.16 no.1
• /
• pp.63-69
• /
• 2012

This paper presents a method to derive the theoretical requirements for the development of a 400 mm optical window that transmits dual-band wavelengths and had a stable structure. We also present design and fabrication results. Among the required specifications, the surface figure error was defined by the transmitted wavefront deformation (TWD), ${\lambda}$/15 rms at 632.8 nm. This value was derived by estimating the predicted performances with respect to five independent items that could cause system performance degradation and then calculating the required wavefront error (WFE) to satisfy the performance goals. We measured the image resolution at each performance level to trace and verify the requirements. The article also describes a design optimization process that could minimize the weight and volume of the optical window attached to the payload securing the FOV of the camera. In addition, we accurately measured the deformation that occurred in the series of fabrication steps including processing, coating, assembly, bonding and bolting, and investigated the effects by comparing them to the results of a simulation performed in advance to derive the predicted performance.

• Journal of Korea Society of Digital Industry and Information Management
• /
• v.8 no.2
• /
• pp.131-143
• /
• 2012

This paper introduces the double-density discrete wavelet transform using 3 direction separable processing method, which is a discrete wavelet transform that combines the double-density discrete wavelet transform and quincunx sampling method, each of which has its own characteristics and advantages. The double-density discrete wavelet transform is nearly shift-invariant. But there is room for improvement because not all of the wavelets are directional. That is, although the double-density DWT utilizes more wavelets, some lack a dominant spatial orientation, which prevents them from being able to isolate those directions. The dual-tree discrete wavelet transform has a more computationally efficient approach to shift invariance. Also, the dual-tree discrete wavelet transform gives much better directional selectivity when filtering multidimensional signals. But this transformation has more cost complexity Because it needs eight digital filters. Therefor, we need to hybrid transform which has the more directional selection and the lower cost complexity. A solution to this problem is a the double-density discrete wavelet transform using 3 direction separable processing method. The proposed wavelet transformation services good performance in image and video processing fields.

• Proceedings of the IEEK Conference
• /
• 2006.06a
• /
• pp.503-504
• /
• 2006

This paper presents the design and fabrication of a LC voltage-controlled oscillator (VCO) using 1-poly 6-metal mixed signal CMOS process. To obtain the high-quality factor inductor in LC resonator, patterned-ground shields (PGS) is placed under the symmetric inductor to reduce the effect from image current of resistive Si substrate. Moreover, due to the incapability of using thick top metal layer of which the thickness is over $2{\mu}m$, as used in many RF CMOS process, the structure of dual-metal layer in which we make electrically short circuit between the top metal and the next metal below it by a great number of via materials along the metal traces is adopted. The circuit operated from 2.63 GHz to 3.09 GHz tuned by accumulation-mode MOS varactor. The corresponding tuning range was 460 MHz. The measured phase noise was -115 dBc/Hz @ 1MHz offset at 2.63 GHz carrier frequency and the current consumption and the corresponding power consumption were about 2.6 mA and 4.68 mW respectively.

• International Journal of CAD/CAM
• /
• v.11 no.1
• /
• pp.18-26
• /
• 2011

This paper proposes a new solid-shape modeling system based on a lusterware-image illustration. The proposed method reconstructs a three dimensional solid shape from a set of rough sketches that are typically drawn in the early stages of the design process. The sketches do not have to be strictly accurate, and this tolerance to the roughness of the input sketches is one of the major advantages of the proposed method. The proposed system creates an initial shape based on the silhouette of the input lusterware-images. Then the user can edit the initial shape with intuitive cutting and dishing-up operations, which are based on sketching user interface. To achieve the goal, the system retains the geometric model with two representations: a point-set data and a volume data. This dual data structure allows the program to create an initial shape from the input images with little computational cost, and the user can apply cutting and dishing-up operations without substantially increasing computational and memory requirements. In this research, we have tested the proposed system by reconstructing solid models of some mechanical parts from rough sketches. The experimental results indicate that the proposed method is useful for the prototyping of a solid shape.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.18 no.1
• /
• pp.15-29
• /
• 2024

A fetoscope is an optical endoscope, which is often applied in fetoscopic laser photocoagulation to treat twin-to-twin transfusion syndrome. In an operation, the clinician needs to observe the abnormal placental vessels through the endoscope, so as to guide the operation. However, low-quality imaging and narrow field of view of the fetoscope increase the difficulty of the operation. Introducing an accurate placental vessel segmentation of fetoscopic images can assist the fetoscopic laser photocoagulation and help identify the abnormal vessels. This study proposes a method to solve the above problems. A novel encoder-decoder network with a dual-path structure is proposed to segment the placental vessels in fetoscopic images. In particular, we introduce a channel attention mechanism and a continuous convolution structure to obtain multi-scale features with their weights. Moreover, a switching connection is inserted between the corresponding blocks of the two paths to strengthen their relationship. According to the results of a set of blood vessel segmentation experiments conducted on a public fetoscopic image dataset, our method has achieved higher scores than the current mainstream segmentation methods, raising the dice similarity coefficient, intersection over union, and pixel accuracy by 5.80%, 8.39% and 0.62%, respectively.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.10 no.1
• /
• pp.158-164
• /
• 2009

The real time security system using web camera shall correspond in commensurate with it in real time through classifying moving object and analyzing the behavior. But, as to the detection of moving object in real time image through a camera, it is difficult to detect movement correctly according to the change of unnecessary noises, lighting conditions and screened phenomenon. This paper proposes real time security system by dual camera and ultrasonic sensor, a method of advanced detection in order to detect correct movement of specific object. That is, we could improve the tracing characteristics by using ultrasonic sensor as measurement factor of changed position and verify through experiments that the information interchanged between camera upwards and in front of it have effect on tracing a specific object continuously. The results of the experiment show that recognition rate of object was 97.4% and the correct tracing could be done lastingly in a phenomena of screening object.

• Journal of the Korea Fashion and Costume Design Association
• /
• v.17 no.3
• /
• pp.135-148
• /
• 2015

The purpose of this study was to research into TV makeup types focusing on Korean aesthetics with recognizing importance of Korean thought amid what the global interest pays attention to South Korea thanks to the recent influence of Hanryu(Korean wave). The following are summary and result of this study. The non-dual beauty is an integrative concept of including ambivalence on the categorical difference in sex, culture and class by transcending the extreme binary division. As this is what reflected the open thought of pursuing balance of yin and yang in Korean people, it is being shown in the form of transcendence and far-outness through TV entertainment programs. Transcendence coexisted with maintaining relative relationship beyond separation in sex. Far-outness was pursuing free sensitivity immanent in the non-separated thought of transcending the past, the present and future. This trend is being indicated similarly to a change in its paradigm from separation to convergence these days. Accordingly, even makeup was showing similarity to the recent trend with being paid attention to the makeup of focusing on identity in sex and to the makeup of being coexisted the past, the present and future with the aspect of being mixed space time. The Korean aesthetics has been feeble in its influence compared to the Oriental image of focusing on China and Japan in the meantime. Hence, this study can be said to have significance in a sense of being the first consideration that compared TV makeup types of focusing on Korean aesthetics and suggested its developmental direction.

• 제어로봇시스템학회:학술대회논문집
• /
• 1992.10b
• /
• pp.382-387
• /
• 1992

In order to obtain desired arc welding performance, we already developed an arc welding robot system that enabled coordinated motions of dual arm robots. In this system one robot arm holds a welding target as a positioning device, and the other robot moves the welding torch. Concerning to such a dual arm robot system, the positioning accuracy of robots is one important problem, since nowadays conventional industrial robots unfortunately don't have enough absolute accuracy in position. In order to cope with this problem, our robot system employed teaching playback method, where absolute error are compensated by the operator's visual feedback. Due to this system, an ideal arc welding considering the posture of the welding target and the directions of the gravity has become possible. Another problem still remains, while we developed an original teaching method of the dual arm robots with coordinated motions. The problem is that manual teaching tasks are still tedious since they need fine movements with intensive attentions. Therefore, we developed a 3-dimensional vision guided robot control method for our welding robot system with coordinated motions. In this paper we show our 3-dimensional vision sensor to guide our arc welding robot system with coordinated motions. A sensing device is compactly designed and is mounted on the tip of the arc welding robot. The sensor detects the 3-dimensional shape of groove on the target work which needs to be weld. And the welding robot is controlled to trace the grooves with accuracy. The principle of the 3-dimensional measurement is depend on the slit-ray projection method. In order to realize a slit-ray projection method, two laser slit-ray projectors and one CCD TV camera are compactly mounted. Tactful image processing enabled 3-dimensional data processing without suffering from disturbance lights. The 3-dimensional information of the target groove is combined with the rough teaching data they are given by the operator in advance. Therefore, the teaching tasks are simplified

• Journal of Biomedical Engineering Research
• /
• v.27 no.5
• /
• pp.224-228
• /
• 2006

The object of this study is to investigate the potential of dual-electrospun polymer based structure for vascular tissue engineering, especially for the medium or small sue blood vessels. Polyurethane(PU), which is known to be biocompatible in this area, was electrospun with poly(ethylene oxide) (PEO). Concentration of PU was fixed at 20wt%, while that of PEO was set from 15 to 35wt%. Morphological features were observed by SEM image and measurement of porosity and cellular responses were tested before and after extracting PEO from the hybrid scaffolds by immersing the scaffolds into distilled water. The diameter of PEO fibers were ranged from 200nm to 500nm. The lower concentration of PEO tended to show beads. The porosity of the scaffolds after extracting PEO was highly increased with higher concentration of PEO as expected. Also, higher proliferation rate of smooth muscle cells was observed at higher concentration of PEO than at the lower concentration and without PEO. As conclusions, this dual electrospinning technique combined with PU and PEO is expected to overcome the current barrier of cell penetration by providing more space for cells to proliferation.