• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.9
• /
• pp.2882-2903
• /
• 2022

Image super-resolution (SR) processing is of great value in the fields of digital image processing, intelligent security, film and television production and so on. This paper proposed a densely connected deep learning network based on cascade architecture, which can be used to solve the problem of super-resolution in the field of image quality enhancement. We proposed a more efficient residual scaling dense block (RSDB) and the multi-channel cascade architecture to realize more efficient feature reuse. Also we proposed a hybrid loss function based on L₁ error and L_∞ error to achieve better L_∞ error performance. The experimental results show that the overall performance of the network is effectively improved on cascade architecture and residual scaling. Compared with the residual dense net (RDN), the PSNR / SSIM of the new method is improved by 2.24% / 1.44% respectively, and the L_∞ performance is improved by 3.64%. It shows that the cascade connection and residual scaling method can effectively realize feature reuse, improving the residual convergence speed and learning efficiency of our network. The L_∞ performance is improved by 11.09% with only a minimal loses of 1.14% / 0.60% on PSNR / SSIM performance after adopting the new loss function. That is to say, the L_∞ performance can be improved greatly on the new loss function with a minor loss of PSNR / SSIM performance, which is of great value in L_∞ error sensitive tasks.

• Journal of Institute of Control, Robotics and Systems
• /
• v.18 no.1
• /
• pp.47-53
• /
• 2012

In cellular geo-location systems, positioning performance is influenced by hearability of receivers. Hearability can be enhanced by using long integration at the receiver. When unknown residual frequency remains in baseband signals, however, the coherent integration loss increases as the residual frequency becomes larger. Consequently, length of coherent integration is determined by the residual frequency. By precise estimation and compensation of the residual frequency, integration length can be enlarged. This paper presents a residual frequency estimator for WiBro geo-location and analyzes its performance in multipath environment. By computer simulation, an optimal receiver structure to enhance the hearability of WiBro geo-location is proposed.

• Environmental Engineering Research
• /
• v.10 no.4
• /
• pp.155-164
• /
• 2005

This study investigated the possibility of reducing the residual aluminum (Al) in the treated water using polymers. Two raw waters (lake and river water) and three kinds of polymers (coagulant, flocculant, and filtration aids) were used for this purpose. This study found that coagulation at the high dose did not necessarily lead to the high concentration of the residual Al in the treated water. The coagulation efficacy was found more important in determining the residual Al than the coagulant dose. The use of a polymer enhanced the removal of turbidity as well as the residual Al. The coagulant aid removed the dissolved Al as well as the particulate Al by helping the coagulation and the solid-liquid separation. The flocculant aid and the filtration aid preferentially removed the particulate Al while helping the solid-liquid separation. The filtration aid reduced the residual Al substantially more effectively than the flocculant aid. The polyamine-based coagulant aid (FL) showed the better performance in reducing the residual Al and turbidity than DADMAC (WT). The cationic flocculant aid with weak charge density and the medium molecular weight (SC-020) showed the best performance in reducing the residual Al.

• Journal of Institute of Control, Robotics and Systems
• /
• v.21 no.7
• /
• pp.684-690
• /
• 2015

This paper presents a modified residual-based EKF (Extended Kalman Filter) for performance improvement of indoor positioning using WiFi RSSI (Received Signal Strength Indicator) measurement. Radio signal strength in indoor environments may have irregular attenuation characteristics due to obstacles such as walls, furniture, etc. Therefore, the performance of the RSSI-based positioning with the conventional trilateration method or Kalman filter is insufficient to provide location-based accurate information services. In order to enhance the performance of indoor positioning, in this paper, error analysis of the distance calculated by using the WiFi RSSI measurement is performed based on the radio propagation model. Then, an IARM (Irregularly Attenuated RSSI Measurement) error is defined. Also, it shows that the IARM error is included in the residual of the positioning filter. The IARM error is always positive. So, it is presented that the IARM error can be estimated by taking the absolute value of the residual. Consequently, accurate positioning can be achieved based on the IEM (IARM Error Mitigated) EKF with the residual modified by using the estimated IARM error. The performance of the presented IEM EKF is verified experimentally.

• Journal of Institute of Control, Robotics and Systems
• /
• v.16 no.7
• /
• pp.720-725
• /
• 2010

In AGPS method, user position can be obtained even in the shadow region by improving signal sensitivity. A hybrid long integration scheme employing both coherent and non-coherent integration method is commonly used in AGPS receivers. Because coherent loss increases as residual frequency become large, residual frequency should be minimized to maximize coherent integration gain. This paper presents performance analysis of residual frequency estimator using FFT in fine-time assistance AGPS method. Considering the hardware complexity and the estimation accuracy, optimal length of FFT is proposed for GPS L1 C/A signal. Signal sensitivity for estimating the residual frequency is also analysed. By field experimental results, it is found that the residual frequency can be successfully estimated using 1 second snap-shot data when GPS signal strength is larger than -150 dBm and its RMS error is 3Hz.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.18 no.6
• /
• pp.113-119
• /
• 2018

This paper relates the AM-NM-MMA algorithm which possible to adaptive equalization performance improvement using the adaptive modulus instead of fixed modulus in the NM-MMA algorithm. The NM-MMA emerged for the tradeoff the MMA and SE-MMA algorithm characteristics, the MMA provides the less residual values in the steady state and have a slow convergence rate, the SE-MMA provides the fast convergence rate and increae the risidual values in the steady state. But the fixed modulus can not give the zero residual values in the perfect equalization state and eqaulization performance were degrade, the adaptive modulus was applied in order to reducing the residual values, and its improved performance were confirmed by simulation. For this, the equalizer output constellation, residual isi, MD, MSE, SER were used for performance index. As a result of computer simulation, the AM-NM-MMA has more good performance in equalizer output signal constellation, residual isi, MD, MSE than the NM-MMA, but not in SER performance.

• Earthquakes and Structures
• /
• v.10 no.2
• /
• pp.409-428
• /
• 2016

To realise the full benefits of a self-centering seismic resilient system, the designer must ensure that the entire structure does indeed re-center following an earthquake. The idealised flag-shaped hysteresis response that is often used to define the cyclic behaviour of self-centering concrete systems seldom exists and the residual drift of a building subjected to an earthquake is dependent on the realistic cyclic hysteresis response as well as the dynamic loading history. Current methods that are used to ensure that re-centering is achieved during the design of self-centering concrete systems are presented, and a series of cyclic analyses are used to demonstrate the flaws in these current procedures, even when idealised hysteresis models were used. Furthermore, results are presented for 350 time-history analyses that were performed to investigate the expected residual drift of an example self-centering concrete wall system during an earthquake. Based upon the results of these time-history analyses it was concluded that due to dynamic shake-down the residual drifts at the conclusion of the ground motion were significantly less than the maximum possible residual drifts that were observed from the cyclic hysteresis response, and were below acceptable residual drift performance limits established for seismic resilient structures. To estimate the effect of the dynamic shakedown, a residual drift ratio was defined that can be implemented during the design process to ensure that residual drift performance targets are achieved for self-centering concrete wall systems.

• Journal of Information Processing Systems
• /
• v.16 no.6
• /
• pp.1479-1494
• /
• 2020

As an important part of image processing, image demosaicking has been widely researched. It is especially necessary to propose an efficient interpolation algorithm with good visual quality and performance. To improve the limitations of residual interpolation (RI), based on RI algorithm, minimalized-Laplacian RI (MLRI), and iterative RI (IRI), this paper focuses on adaptive RI (ARI) and proposes an improved ARI (IARI) algorithm which obtains more distinct R, G, and B colors in the images. The proposed scheme fully considers the brightness information and edge information of the image. Since the ARI algorithm is not completely adaptive, IARI algorithm executes ARI algorithm twice on R and B components according to the directional difference, which surely achieves an adaptive algorithm for all color components. Experimental results show that the improved method has better performance than other four existing methods both in subjective assessment and objective assessment, especially in the complex edge area and color brightness recovery.

• Proceedings of the Korean Institute of Navigation and Port Research Conference
• /
• 2002.11a
• /
• pp.13-19
• /
• 2002

Design and implementation methodologies on the performance evaluation system of two-channel ring-core Flux-Gate Compass (FG-Compass) are described, with evaluation procedures and methods based on the polynomial regression models. Performance evaluation system is consists of a step motor driving unit, a bearing transmitting unit and, evaluation programs using polynomial regression formulae. Through performance evaluation tests, total residual deviation tests, total residual deviation of $\pm$4$^{\circ}$ and eigen residual deviation of $\pm$2$^{\circ}$ are obtained from the FG-Compass. The result is more accurate values than the typical FG-Compass with eigen residual deviation of $\pm$4$^{\circ}$ and is provide a possibility to develop a high performance FG-Compass. In addition, the design methodology of a smart FG-Compass with the self estimation and correction of residual deviations is also discussed.

• International Journal of Computer Science & Network Security
• /
• v.21 no.9
• /
• pp.203-211
• /
• 2021

Image tampering detection and localization have become an active area of research in the field of digital image forensics in recent times. This is due to the widespread of malicious image tampering. This study presents a new method for image tampering detection and localization that combines the advantages of dilated convolution, residual network, and UNET Architecture. Using the UNET architecture as a backbone, we built the proposed network from two kinds of residual units, one for the encoder path and the other for the decoder path. The residual units help to speed up the training process and facilitate information propagation between the lower layers and the higher layers which are often difficult to train. To capture global image tampering artifacts and reduce the computational burden of the proposed method, we enlarge the receptive field size of the convolutional kernels by adopting dilated convolutions in the residual units used in building the proposed network. In contrast to existing deep learning methods, having a large number of layers, many network parameters, and often difficult to train, the proposed method can achieve excellent performance with a fewer number of parameters and less computational cost. To test the performance of the proposed method, we evaluate its performance in the context of four benchmark image forensics datasets. Experimental results show that the proposed method outperforms existing methods and could be potentially used to enhance image tampering detection and localization.