• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.12
• /
• pp.3364-3382
• /
• 2023

Remote sensing image segmentation plays an important role in realizing intelligent city construction. The current mainstream segmentation networks effectively improve the segmentation effect of remote sensing images by deeply mining the rich texture and semantic features of images. But there are still some problems such as rough results of small target region segmentation and poor edge contour segmentation. To overcome these three challenges, we propose an improved semantic segmentation model, referred to as MRU-Net, which adopts the U-Net architecture as its backbone. Firstly, the convolutional layer is replaced by BasicBlock structure in U-Net network to extract features, then the activation function is replaced to reduce the computational load of model in the network. Secondly, a hybrid multi-scale recognition module is added in the encoder to improve the accuracy of image segmentation of small targets and edge parts. Finally, test on Massachusetts Buildings Dataset and WHU Dataset the experimental results show that compared with the original network the ACC, mIoU and F1 value are improved, and the imposed network shows good robustness and portability in different datasets.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.6 no.9
• /
• pp.1998-2016
• /
• 2012

Reliable spectrum sensing algorithm is a fundamental component in cognitive radio. In this paper, a non-cooperative spectrum sensing algorithm which needs only one cognitive radio node named CORDIC (Coordinate Rotation Digital Computer) Jacobi based method is proposed. The algorithm computes the eigenvalues of the sampled covariance of received signal mainly by shift and additional operations, which is suitable for hardware implementation. Based the latest random matrix theory (RMT) about the distribution of the limiting maximum and minimum eigenvalue ratio, the relationship between the probability of false alarm and the decision threshold is derived. Simulations and discussions show the method is effective. Real captured digital television (DTV) signals and Universal Software Radio Peripheral (USRP) are also employed to evaluate the performance of the algorithm, which prove the proposed algorithm can be applied in practical spectrum sensing applications.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.10 no.3
• /
• pp.43-50
• /
• 2010

DDC(Digital Down Converter) is a conversion technology to decimate to a lower sampling rate and DDC for the future development of communications technology has the necessary skills. So, it has been recognized in the wireless and the SDR(Software Defined Radio) system as essential components. In addition, research is underway on spectrum sensing for efficient communications environment due to the shortage of frequency resources. In this paper, the DDC systems were analyzed for CIC(Cascaded Integrator Comb) Filter, WDF(Wave Digital Filter), SRC(Sample Rate Conversion) each module. Moreover, we proposed a linkage effectively between DDC system and Spectrum Sensing for improve the efficiency of use of frequency by computer simulations. The simulation results of the DDC system was applied to the spectrum sensing capabilities. Also, performance and complexity of the results were derived and proposed system was the result of the check.

• Proceedings of the KSRS Conference
• /
• 2002.10a
• /
• pp.739-744
• /
• 2002

Recently we have discovered that sediments should be separated from lithosphere, and soil should be separated from biosphere, both sediment and soil will be mixed sediments-soil-sphere (Seso-sphere), which is using particulate mechanics to be solved. Erosion and sediment both are moving by particulate matter with water or wind. But ancient sediments will be erosion same to soil. Nowadays, real soil has already reduced much more. Many places have only remained sediments that have ploughed artificial farming layer. Thus it means sediments-soil-sphere. This paper discusses sediments-soil-sphere erosion modeling. In fact sediments-soil-sphere erosion is including water erosion, wind erosion, melt-water erosion, gravitational water erosion, and mixed erosion. We have established geographical remote sensing information modeling (RSIM) for different erosion that was using remote sensing digital images with geographical ground truth water stations and meteorological observatories data by remote sensing digital images processing and geographical information system (GIS). All of those RSIM will be a geographical multidimensional gray non-linear equation using mathematics equation (non-dimension analysis) and mathematics statistics. The mixed erosion equation is more complex that is a geographical polynomial gray non-linear equation that must use time-space fuzzy condition equations to be solved. RSIM is digital image modeling that has separated physical factors and geographical parameters. There are a lot of geographical analogous criterions that are non-dimensional factor groups. The geographical RSIM could be automatic to change them analogous criterions to be fixed difference scale maps. For example, if smaller scale maps (1:1000 000) that then will be one or two analogous criterions and if larger scale map (1:10 000) that then will be four or five analogous criterions. And the geographical parameters that are including coefficient and indexes will change too with images. The geographical RSIM has higher precision more than mathematics modeling even mathematical equation or mathematical statistics modeling.

• Proceedings of the Korean Institute of Navigation and Port Research Conference
• /
• 2004.08a
• /
• pp.29-36
• /
• 2004

The paper describes recent experimental results on the development of Digital Magnetic Compass (DMC), which can provide smart automatic correction functions to the magnetic interferences. The design methodology of magnetic sensing circuit with ring-core fluxgate sensor is represented. The performance results of the sensing circuits are discussed with error analysis by polynomial regressions. As test results, the sensing circuit filtered only the second harmonic signal that is proportional to the direction of earth's magnetic field, and it leads to the obtainment of bearing information. In addition, the total residual errors of DMC can be analyzed by the adoption of polynomial regressions. It shown that the possibility of high precise DMC, in the future.

• Korean Institute of Interior Design Journal
• /
• v.23 no.2
• /
• pp.80-89
• /
• 2014

0 and 1, two digits are principle and formality of digital media. Those principle and formality dwell in new media paradigm. They are founded on emergence of new types of space. As describing the relationship originating from the concepts of digital media paradigm and space and discussing their characteristics, this paper presents the grounds for a theoretical argument for formation of new space. The concept of virtuality which mimics and represents the real, non-linearity which resembles human system of thinking and sensing, and interactivity connecting the communication theory of human and machine, their properties form the important basis for emergence of a new aspect of space. The concepts of digital media paradigm divide a type of space into two properties. The first is a inner modality. The second is outer modality. Dynamic space implicates inner modality that humans internally recognize the sense of space through their body and sensing organs. Space is not sensed but sensing. Space has an organic nature through object's interaction. Space has an outer nature that is physically variable. Finally, space has properties of modularity that changes structure of space. For empirical basis of an argument, new media art that has architectural form and those spatial characteristics were compared and analyzed. This conceptual discussion for space which the formality of digital media is applied will be an important foundation for create space design.

• Korean Journal of Remote Sensing
• /
• v.2 no.2
• /
• pp.69-77
• /
• 1986

This study was emphasized on the land use classification by Remote Sensing Technique. Land cover maps about the major urbans, Seoul, Tajeon regions, its of each classified classes were extracted by use of Landsat MSS Data and Digital Image Processing System. From the results of this study, it was proved that land use classification by Remote Sensing technique could be used to obtain fully fruitful Results.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.54 no.6
• /
• pp.134-139
• /
• 2017

In this paper, a novel torque control scheme is proposed and implemented to handle the torque level of ebike precisely. By adopting moving average filters to eliminate throttle noise, ebike driver could control throttle level on wide span of 256 steps. Designed controller is plugged into ebike and tested to demonstrate it's linear control incomparable to conventional cadence sensing controller.

• Proceedings of the KSRS Conference
• /
• 2003.11a
• /
• pp.885-887
• /
• 2003

The purpose of this investigation is to generate true orthophotos from high resolution satellite images. The major works of this research include 4 parts: (1) determination of orientation parameters, (2) generating traditional orthophotos using terrain model, (3) relief correction for buildings, and (4) process for hidden areas. To determine the position of satellites, we correct the onboard orientation parameters to fine tune the orbit. In the generation of traditional orthophotos, we employ orientation parameters and digital terrain model(DTM) to rectify tilt displacements and relief displacements for terrain. We, then, compute relief displacements for buildings with digital building model (DBM). To avoid double mapping, we detect hidden areas. Due to the satellite’s small field of view, an efficient method for the detection of hidden areas and building rectification will be proposed in this paper. Test areas cover the city of Kaohsiung in southern Taiwan. Test images are from the QuickBird satellite.

• Korean Journal of Remote Sensing
• /
• v.10 no.1
• /
• pp.15-29
• /
• 1994

Radiometric correction refers to variations in the data that are not caused by the object or scene being scanned. These variations can be caused by differing sensitivities of the detectors of the sensing system, malfunctioning detectors, or atmospheric interference. Radiometric corrections can be applied to correct for these variations, such as for differing sensitivities of detectors (causing striped image), for detectors (resulting in pixels with digital values of zero), or to correct for atmospheric bias due to scattering of radiation. This paper discussed and illustrated some of the important principles of the radiometric correction methods.