• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.44 no.1
• /
• pp.85-93
• /
• 2024

The monitoring of environmental information (e.g. noise, dust, vibration, temperature, humidity) is crucial to the safe and sustainable operation of a construction site. However, commercial sensors exhibit certain drawbacks when applied on-site. First, the installation cost is prohibitively high. Second, these sensors have been engineered without considering the rugged and harsh conditions of a construction site, resulting in error-prone sensing. Third, construction sites are compelled to allocate additional resources in terms of manpower, expenses, and physical spaces to accommodate individual sensors. This research developed an integrated sensing module to measure the environmental information in construction site. The sensing module slashes the installation cost to 3.3%, is robust enough to harsh and outdoor sites, and consolidates multiple sensors into a single unit. The sensing module also supports GPS, LTE, and real-time sensing. The evaluation showed remarkable results including 97.5% accuracy and 99.9% precision in noise measurement, an 89.7% accuracy in dust measurement, and a 93.5% reliability in data transmission. This research empowers the collection of substantial volumes and high-quality environmental data from construction sites, providing invaluable support to decision-making process. These encompass objective regulatory compliance checking, simulations of environmental data dispersion, and the development of environmental mitigation strategies.

• Journal of Korea Multimedia Society
• /
• v.16 no.2
• /
• pp.169-179
• /
• 2013

Change detection algorithms take two image frames and return the locations of newly introduced objects which cause differences between the images. This paper presents a new change detection method, which classifies intensity changes due to introduced objects, reflected light and shadow from the objects to their neighborhood, and the noise, and exactly localizes the introduced objects. For classification and localization, first we analyze the histogram of the intensity difference between two images, and estimate multiple threshold values. Second we estimate candidate object boundaries using the gradient difference between two images. Using those threshold values and candidate object boundaries, we segment the frame difference image into multiple regions. Finally we classify whether each region belongs to the introduced objects or not using textures in the region. Experiments show that the proposed method exactly localizes the objects in various scenes with different lighting.

• Journal of Power Electronics
• /
• v.15 no.3
• /
• pp.730-740
• /
• 2015

This paper presents model reference adaptive system speed estimators based on Type-1 and Type-2 fuzzy logic controllers for the speed sensorless direct torque and flux control of an induction motor drive (IMD) using space vector pulse width modulation. A Type-1 fuzzy logic controller (T1FLC) based adaptation mechanism scheme is initially presented to achieve high performance sensorless drive in both transient as well as in steady-state conditions. However, the Type-1 fuzzy sets are certain and cannot work effectively when a higher degree of uncertainties occurs in the system, which can be caused by sudden changes in speed or different load disturbances and, process noise. Therefore, a new Type-2 FLC (T2FLC) - based adaptation mechanism scheme is proposed to better handle the higher degree of uncertainties, improve the performance, and is also robust to different load torque and sudden changes in speed conditions. The detailed performance of different adaptation mechanism schemes are performed in a MATLAB/Simulink environment with a speed sensor and sensorless modes of operation when an IMD is operates under different operating conditions, such as no-load, load, and sudden changes in speed. To validate the different control approaches, the system is also implemented on a real-time system, and adequate results are reported for its validation.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.8 no.2
• /
• pp.483-503
• /
• 2014

In this paper, human action recognition using pyramid histograms of oriented gradients and collaborative multi-task learning is proposed. First, we accumulate global activities and construct motion history image (MHI) for both RGB and depth channels respectively to encode the dynamics of one action in different modalities, and then different action descriptors are extracted from depth and RGB MHI to represent global textual and structural characteristics of these actions. Specially, average value in hierarchical block, GIST and pyramid histograms of oriented gradients descriptors are employed to represent human motion. To demonstrate the superiority of the proposed method, we evaluate them by KNN, SVM with linear and RBF kernels, SRC and CRC models on DHA dataset, the well-known dataset for human action recognition. Large scale experimental results show our descriptors are robust, stable and efficient, and outperform the state-of-the-art methods. In addition, we investigate the performance of our descriptors further by combining these descriptors on DHA dataset, and observe that the performances of combined descriptors are much better than just using only sole descriptor. With multimodal features, we also propose a collaborative multi-task learning method for model learning and inference based on transfer learning theory. The main contributions lie in four aspects: 1) the proposed encoding the scheme can filter the stationary part of human body and reduce noise interference; 2) different kind of features and models are assessed, and the neighbor gradients information and pyramid layers are very helpful for representing these actions; 3) The proposed model can fuse the features from different modalities regardless of the sensor types, the ranges of the value, and the dimensions of different features; 4) The latent common knowledge among different modalities can be discovered by transfer learning to boost the performance.

• 제어로봇시스템학회:학술대회논문집
• /
• 2004.08a
• /
• pp.243-247
• /
• 2004

The dynamometers which had made in a long time ago could not control the input/output quantity of water minutely and was sensitive to a noise since it was controlled by an analog control method. Therefore, a fully digital controlled system was urgently required to be robust against various noises. In this paper, the new system which can control the amount of circulated water in dynamometer was developed. This system is consisted of an industrial digital type controller and a servo motor. The industrial PLC was used as a main controller for the developed system, and the actuator and servo motor were used to control the inlet and outlet valve independently. The torque signal of load cell was fed back to the main controller to regulate the diesel engines load. Generally, an input/output valve position of the old dynamometer was fixed with a proper situation for an engine output test and the torque was changed according to the time interval. However, the torque value for the dynamometer could not be constantly kept because of the variation of the input water flow and fluid characteristic. Therefore, the automatic control of an inlet and outlet valve should be performed to keep the constant torque. So, the PID control method was applied to solve this problem. Also, the development of a web-based remote control system was described in this paper. This software will give us the convenience of operation, the more efficient operations, and the reduced operator workload for operation of the dynamometer. The application results of the system have been verified at actual diesel engine field.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.41 no.3
• /
• pp.73-81
• /
• 2004

To extract the creditable features in fingerprint image, many people use the thinning algorithm that has a very important position in the preprocessing. In this paper, we propose the robust parallel thinning algorithm that can preserve the connectivity of the binarized fingerprint image, make the thinnest skeleton with 1-pixel width and get near to the medial axis extremely. The proposed thinning method repeats three sub-iterations. The first sub-iteration takes off only the outer boundary pixel by using the interior points. To extract the one side skeletons, the second sub-iteration finds the skeletons with 2-pixel width. The third sub-iteration prunes the needless pixels with 2-pixel width existing in the obtained skeletons and then the proposed thinning algorithm has the robustness against the rotation and noise and can make the balanced medial axis. To evaluate the performance of the proposed thinning algorithm we compare with and analyze the previous algorithms.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.49 no.9
• /
• pp.155-163
• /
• 2012

The geometrical and electrical errors of array sensors can severely degrade the performance of array sensor system. Various calibration techniques are developed to alleviate this problem. In this paper, two different calibration methods with respect to location, gain and phase of array sensors are presented. One method applies the first-order Taylor series expansion to approximate the true steering vector from the nominal values of array sensors. Then a set of equations is formed by using the null characteristics of the MUSIC spectrum to estimate errors of location, gain and phase of array sensors. Another method estimates these errors based on the data covariance matrix of pilot sources. From the simulations, it is demonstrated that two calibration algorithms calibrated an array system successfully. In addition to that, Fistas and Manikas's algorithm is more robust against noise than Ng and Lie's one when SNR is from 10dB to 50dB.

• The Journal of the Acoustical Society of Korea
• /
• v.36 no.5
• /
• pp.353-360
• /
• 2017

The single sensor based acoustic reflection control system separates the incident and reflected signals from the single sensor output, and reduces the reflected signal by generating an out-of-phase signal from the incident signal component. In this paper, we propose an optimal filter design method for a single sensor based reflection control system. In the proposed method, it is shown that the optimum control filter design is possible by using the measured impulse responses of the reflection and control paths. The reflection control algorithm based on the proposed optimal filter achieves better performance than the conventional adaptive filter-based algorithm and effectively controls the reflection without the initial convergence time. We performed computer simulations using the signals obtained in a 1-dimensional acoustic duct environment, and from the simulation results, it was confirmed that the proposed optimal filter has robust performance even in noisy environment.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.17 no.6
• /
• pp.521-526
• /
• 2016

This paper suggests a new design that makes use of rotary inertia that can allow autonomous movement of an autonomous colonoscope robot in the colon of a patient as a locomotive mechanism. As commercial colonoscopy causes a lengthy time of pain and discomfort to the patients when colonoscopy patients are reluctant to receive surgery, there is a tendency to avoid the test in the early diagnosis of colorectal cancer. To solve this problem, research has been conducted on the next generation of robotic colonoscopes that can reduce the discomfort and pain by moving autonomously within the colon of the patients. In the driving mechanism utilizing the rotational inertia, a flywheel is driven by a motor to store energy and produce rotational inertia. By the energy stored and released by the flywheel, the stick phenomenon that occurs when the robot is running in the intestine can be overcome effectively. To do this, a controller that can control the velocity of the flywheel and is robust to high frequency noise was designed and implemented. The driving mechanism using the rotational inertia presented here showed that the structure is also effective and the experiment can be run easily compared to another mechanism.

• The Korean Journal of Applied Statistics
• /
• v.22 no.4
• /
• pp.709-721
• /
• 2009

This paper discusses an efficient image segmentation using morphological watershed algorithm that is robust to noise. Morphological image segmentation consists of four steps: image simplification, computation of gradient image and watershed algorithm and region merging. Conventional watershed segmentation exhibits a serious weakness for over-segmentation of images. In this paper we present a morphological edge detection methods for detecting edges under noisy condition and apply our watershed algorithm to the resulting gradient images and merge regions using Kolmogorov-Smirnov test for eliminating irrelevant regions in the resulting segmented images. Experimental results are analyzed in both qualitative analysis through visual inspection and quantitative analysis with percentage error as well as computational time needed to segment images. The proposed algorithm can efficiently improve segmentation accuracy and significantly reduce the speed of computational time.