Browsing by Author "Chen, Kuan-Lin"

Now showing 1 - 5 of 5

APPLYING GENERALIZED WEIGHTED MEAN AGGREGATION TO IMPULSIVE NOISE REMOVAL OF IMAGES
(2014)
Chen, Kuan-Lin
;
Chang, Jyh-Yeong
;
電機工程學系
;
Department of Electrical and Computer Engineering
In this paper, we apply generalized weighted mean to construct interval-valued fuzzy relations for grayscale image impulse noise detection and correction. First, we employ two weighting parameters and perform the weighted mean aggregation for the central pixel and its eight neighbor pixels in a 3x3 sliding window across the image. Then, to counter the over-weighting of a big difference term, we apply a saturation threshold transfer function to these eight pixel difference values. Finally, the image noise map is obtained through a threshold operation on the cumulative differences. To decrease the noise detection error, weighting parameters of the mean can be learned by the gradient method caste in discrete formulation. Moreover, to get higher PSNR in the corrected image, we have experienced from the training that we will select weight of 20 for noise rate smaller than 20% and 50 for noise rate greater than 20%, on erroneous noisy than that on the erroneous non-noise pixel. By the experiment, we have shown that the integration of interval-valued fuzzy relations with the weighted mean aggregation algorithm can effectively detect the image noise pixels and then correct them thereafter.
Deep learning and control algorithms of direct perception for autonomous driving
(1970-01-01)
Lee, Der-Hau
;
Chen, Kuan-Lin
;
Liou, Kuan-Han
;
Liu, Chang-Lun
;
Liu, Jinn-Liang
We propose an end-to-end machine learning model that integrates multi-task (MT) learning, convolutional neural networks (CNNs), and control algorithms to achieve efficient inference and stable driving for self-driving cars. The CNN-MT model can simultaneously perform regression and classification tasks for estimating perception indicators and driving decisions, respectively, based on the direct perception paradigm of autonomous driving. The model can also be used to evaluate the inference efficiency and driving stability of different CNNs on the metrics of CNN's size, complexity, accuracy, processing speed, and collision number, respectively, in a dynamic traffic. We also propose new algorithms for controllers to drive a car using the indicators and its short-range sensory data to avoid collisions in real-time testing. We collect a set of images from a camera of The Open Racing Car Simulator in various driving scenarios, train the model using this dataset, test it in unseen traffics, and find that it outperforms earlier models in highway traffic. The stability of end-to-end learning and self driving depends crucially on the dynamic interplay between CNN and control algorithms. The source code and data of this work are available on our website, which can be used as a simulation platform to evaluate different learning models on equal footing and quantify collisions precisely for further studies on autonomous driving.
Effect of Reconstruction Scheme on the Plasma Fluid Modeling With Original and Reformulated Ion-Related Modeling Equations Using HLL Flux Scheme
(2019-02-01)
Chen, Kuan-Lin
;
Tseng, Meng-Fan
;
Gu, Bi-Ren
;
Sharma, Sarveshwar
;
Wu, Jong-Shinn
In this paper, we would like to investigate the effect of spatial reconstruction schemes on the plasma fluid model with original and reformulated ion-related modeling equations by using the Harten-Lax-van Leer (HLL) flux scheme. The results based on the 1-D fluid model equations for electropositive plasma show that the appearance of unphysical oscillations near-sheath edges strongly depends on the reconstruction schemes (e.g., kappa = 1/2, 1/3, and first-order scheme) when the original ion equations utilize the incorrect ion sound speed. The unphysical oscillations disappear no matter what the spatial reconstruction scheme is applied if the reformulated ion equations with correct ion sound speed are used instead. For electronegative plasma, the kappa reconstruction scheme causes numerical instability when using either the incorrect or correct numerical ion sound speed, but the min-mod reconstruction scheme does not. The numerical results show that for either electropositive or electronegative discharges, both the second-order HLL scheme with min-mod reconstruction and the first-order HLL schemes can produce stable solution, while the accuracy of the former is slightly better than the later. Finally, some examples with an extension to the 2-D-axisymmetric electropositive and electronegative plasmas are demonstrated.
Numerical Investigation on Ion Inertia Force in Low-Temperature Plasma Using Fluid Model Considering Ion Momentum Equation
(2016-12-01)
Chen, Kuan-Lin
;
Tseng, Meng-Fan
;
Gu, Bi-Ren
;
Hung, Chieh-Tsan
;
Wu, Jong-Shinn
In this paper, we would like to numerically investigate the effect of ion inertia force in a typical argon gas discharge using fluid model considering the ion momentum equation, which becomes important under various operating conditions. The proposed self-consistent fluid model consists of the electron continuity equation with the drift-diffusion approximation, the ion continuity equation, the electron energy density equation, the ion momentum equation, and the Poisson equation for electric potential. We studied the plasma properties affected by the inertia force by comparing between the use of momentum equation and drift-diffusion approximation for ion momentum. The results show that the ion inertia force is significant under the low-pressure conditions (<100 mtorr) as expected, e.g., for plasma etching and radio frequency magnetron sputtering plasma. In addition, we also found when the pressure is higher than 100 mtorr, the inertia force becomes important (e.g., more than 15%) in the sheath region when the driving frequency is high enough (e.g., 60 MHz), e.g., for high-frequency plasma-enhanced chemical vapor deposition. Thus, inclusion of inertia force in the modeling of momentum equation becomes necessary under these operating conditions.
Numerical Investigation on Ion Inertia Force in Low-Temperature Plasma Using Fluid Model Considering Ion Momentum Equation
(2016-12)
Chen, Kuan-Lin
;
Tseng, Meng-Fan
;
Gu, Bi-Ren
;
Hung, Chieh-Tsan
;
Wu, Jong-Shinn
In this paper, we would like to numerically investigate the effect of ion inertia force in a typical argon gas discharge using fluid model considering the ion momentum equation, which becomes important under various operating conditions. The proposed self-consistent fluid model consists of the electron continuity equation with the drift-diffusion approximation, the ion continuity equation, the electron energy density equation, the ion momentum equation, and the Poisson equation for electric potential. We studied the plasma properties affected by the inertia force by comparing between the use of momentum equation and drift-diffusion approximation for ion momentum. The results show that the ion inertia force is significant under the low-pressure conditions (<100 mtorr) as expected, e.g., for plasma etching and radio frequency magnetron sputtering plasma. In addition, we also found when the pressure is higher than 100 mtorr, the inertia force becomes important (e.g., more than 15%) in the sheath region when the driving frequency is high enough (e.g., 60 MHz), e.g., for high-frequency plasma-enhanced chemical vapor deposition. Thus, inclusion of inertia force in the modeling of momentum equation becomes necessary under these operating conditions.