• KIPS Transactions on Software and Data Engineering
• /
• v.8 no.9
• /
• pp.379-390
• /
• 2019

In this paper, we propose a novel deep neural network model for Vision-and-Language Navigation (VLN) named LVLN (Landmark-based VLN). In addition to both visual features extracted from input images and linguistic features extracted from the natural language instructions, this model makes use of information about places and landmark objects detected from images. The model also applies a context-based attention mechanism in order to associate each entity mentioned in the instruction, the corresponding region of interest (ROI) in the image, and the corresponding place and landmark object detected from the image with each other. Moreover, in order to improve the success rate of arriving the target goal, the model adopts a progress monitor module for checking substantial approach to the target goal. Conducting experiments with the Matterport3D simulator and the Room-to-Room (R2R) benchmark dataset, we demonstrate high performance of the proposed model.

• International Journal of Computer Science & Network Security
• /
• v.23 no.5
• /
• pp.73-88
• /
• 2023

Brain tumors can also be an abnormal collection or accumulation of cells in the brain that can be life-threatening due to their ability to invade and metastasize to nearby tissues. Accurate diagnosis is critical to the success of treatment planning, and resonant imaging is the primary diagnostic imaging method used to diagnose brain tumors and their extent. Deep learning methods for computer vision applications have shown significant improvements in recent years, primarily due to the undeniable fact that there is a large amount of data on the market to teach models. Therefore, improvements within the model architecture perform better approximations in the monitored configuration. Tumor classification using these deep learning techniques has made great strides by providing reliable, annotated open data sets. Reduce computational effort and learn specific spatial and temporal relationships. This white paper describes transfer models such as the MobileNet model, VGG19 model, InceptionResNetV2 model, Inception model, and DenseNet201 model. The model uses three different optimizers, Adam, SGD, and RMSprop. Finally, the pre-trained MobileNet with RMSprop optimizer is the best model in this paper, with 0.995 accuracies, 0.99 sensitivity, and 1.00 specificity, while at the same time having the lowest computational cost.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.23 no.1
• /
• pp.21-29
• /
• 2003

It is very important to reduce the computational time in estimating the parameters of vision control algorithm for robot's position control in real time. Unfortunately, the batch estimation commonly used requires too murk computational time because it is iteration method. So, the batch estimation has difficulty for robot's position control in real time. On the other hand, the Extended Kalman Filtering(EKF) has many advantages to calculate the parameters of vision system in that it is a simple and efficient recursive procedures. Thus, this study is to develop the EKF algorithm for the robot's vision control in real time. The vision system model used in this study involves six parameters to account for the inner(orientation, focal length etc) and outer (the relative location between robot and camera) parameters of camera. Then, EKF has been first applied to estimate these parameters, and then with these estimated parameters, also to estimate the robot's joint angles used for robot's operation. finally, the practicality of vision control scheme based on the EKF has been experimentally verified by performing the robot's position control.

• Journal of Korea Society of Industrial Information Systems
• /
• v.19 no.4
• /
• pp.41-52
• /
• 2014

In this paper, O-ring size measurement algorithm based on a small machine vision inspection equipment which can replace a expensive and large machine vision inspection equipment is presented. The small machine vision inspection equipment acquires a image from a CCD camera shooting a measurement plane which located on a back light and the proposed size measurement algorithm is apply to the image. For improvement of size measurement accuracy, camera lens distortion correction and perspective distortion correction are conducted by software technique. Consider O-ring's shape, ellipse fitting model is applied. In order to increase the reliability of ellipse fitting, RANSAC algorithm is applied.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.40 no.1
• /
• pp.35-42
• /
• 2012

In this study, a feature-points-tracking algorithm is suggested using a particle filter for vision based navigation system. By applying a dynamic model of the feature point, the tracking performance is increased in high dynamic condition, whereas a conventional KLT (Kanade-Lucas-Tomasi) cannot give a solution. Futhermore, the particle filter is introduced to cope with irregular characteristics of vision data. Post-processing of recorded vision data shows that the tracking performance of suggested algorithm is more robust than that of KLT in high dynamic condition.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.28 no.3
• /
• pp.230-237
• /
• 2004

Lane Sensing techniques based on vision sensors are regarded promising because they require little infrastructure on the highway except clear lane markers. However, they require more intelligent processing algorithms in vehicles to generate the previewed roadway from the vision images. In this paper, a lane sensing algorithm using vision sensors is developed to improve the sensing robustness. The parallel stereo-camera is utilized to regenerate the 3-dimensional road geometry. The lane geometry models are derived such that their parameters represent the road curvature, lateral offset and heading angle, respectively. The parameters of the lane geometry models are estimated by the Kalman filter and utilized to reconstruct the lane geometry in the global coordinate. The inverse perspective mapping from the image plane to the global coordinate considers roll and pitch motions of a vehicle so that the mapping error is minimized during acceleration, braking or steering. The proposed sensing system has been built and implemented on a 1/10-scale model car.

• Journal of the Korean Society of Manufacturing Technology Engineers
• /
• v.23 no.5
• /
• pp.485-497
• /
• 2014

This paper presents the robot vision control schemes based on the Newton-Raphson (N-R) and the Extended Kalman Filter (EKF) methods for the tracking of moving targets. The vision system model used in this study involves the six camera parameters. The difference is that refers to the uncertainty of the camera's orientation and focal length, and refers to the unknown relative position between the camera and the robot. Both N-R and EKF methods are employed towards the estimation of the six camera parameters. Based on the these six parameters estimated using three cameras, the robot's joint angles are computed with respect to the moving targets, using both N-R and EKF methods. The two robot vision control schemes are tested by tracking the moving target experimentally. Given the experimental results, the two robot control schemes are compared in order to evaluate their strengths and weaknesses.

• Journal of Institute of Control, Robotics and Systems
• /
• v.3 no.3
• /
• pp.297-304
• /
• 1997

In stereo vision, camera modeling is very important because the accuracy of the three dimensional locations depends considerably on it. In the existing stereo camera models, two camera planes are located in the same plane or on the optical axis. These camera models cannot be used in the active vision system where it is necessary to obtain two stereo images simultaneously. In this paper, we propose four kinds of stereo camera models for active stereo vision system where focal lengths of the two cameras are different and each camera is able to rotate independently. A single closed form solution is obtained for all models. The influence of the stereo camera model to the field of view, occlusion, and search area used for matching is shown in this paper. And errors due to inaccurate focal length are analyzed and simulation results are shown. It is expected that the three dimensional locations of objects are determined in real time by applying proposed stereo camera models to the active stereo vision system, such as a mobile robot.

• International Journal of Computer Science & Network Security
• /
• v.21 no.9
• /
• pp.179-184
• /
• 2021

In our study we performed a sentiments analysis from the images. For this purpose, we used 153 images that contain: people, animals, buildings, landscapes, cakes and objects that we divided into two categories: images that suggesting a positive or a negative emotion. In order to classify the images using the two categories, we created two models. The SAI-G model was created with Google's AutoML Vision service. The SAI-C model was created on the Clarifai platform. The data were labeled in a preprocessing stage, and for the SAI-C model we created the concepts POSITIVE (POZITIV) AND NEGATIVE (NEGATIV). In order to evaluate the performances of the two models, we used a series of evaluation metrics such as: Precision, Recall, ROC (Receiver Operating Characteristic) curve, Precision-Recall curve, Confusion Matrix, Accuracy Score and Average precision. Precision and Recall for the SAI-G model is 0.875, at a confidence threshold of 0.5, while for the SAI-C model we obtained much lower scores, respectively Precision = 0.727 and Recall = 0.571 for the same confidence threshold. The results indicate a lower classification performance of the SAI-C model compared to the SAI-G model. The exception is the value of Precision for the POSITIVE concept, which is 1,000.

• Proceedings of the Safety Management and Science Conference
• /
• 2012.04a
• /
• pp.139-155
• /
• 2012

A fundamental concepts of business environment changes and the importance of stakeholder's value creation is changing in the business. This study ISO9004: 2009 quality management system of Category 5: Strategy and Policy, Category 10: improvement, innovation and learning (Note) SBK target was to develop a model that is the company's sustained success. Three concepts of the new revision of ISO9004" in response to environmental changes," "learning", "innovation" (Note) SBK applied to the project settings and talent establish long-term vision was to establish the process as the organization's learning content was TDR for the creation of exceptional and innovative programs were introduced. As a result, (Note) SBK three years of continuous business performance indicator has grown dramatically to more than 50% continued success is going to create business models. But 100 years to accomplish the vision, ISO9004 model needs to extends the entire category as a management system to achieve the optimization needed.