• ETRI Journal
• /
• v.46 no.3
• /
• pp.367-378
• /
• 2024

We propose a number-theory-based quantized mathematical optimization scheme for various NP-hard and similar problems. Conventional global optimization schemes, such as simulated and quantum annealing, assume stochastic properties that require multiple attempts. Although our quantization-based optimization proposal also depends on stochastic features (i.e., the white-noise hypothesis), it provides a more reliable optimization performance. Our numerical analysis equates quantization-based optimization to quantum annealing, and its quantization property effectively provides global optimization by decreasing the measure of the level sets associated with the objective function. Consequently, the proposed combinatorial optimization method allows the removal of the acceptance probability used in conventional heuristic algorithms to provide a more effective optimization. Numerical experiments show that the proposed algorithm determines the global optimum in less operational time than conventional schemes.

• International Journal of Computer Science & Network Security
• /
• v.24 no.2
• /
• pp.136-142
• /
• 2024

Haze is a very common phenomenon that degrades or reduces the visibility. It causes various problems where high quality images are required such as traffic and security monitoring. So haze removal from images receives great attention for clear vision. Due to its huge impact, significant advances have been achieved but the task yet remains a challenging one. Recently, different types of deep generative adversarial networks (GAN) are applied to suppress the noise and improve the dehazing performance. But it is unclear how these algorithms would perform on hazy images acquired "in the wild" and how we could gauge the progress in the field. This paper aims to bridge this gap. We present a comprehensive study and experimental evaluation on diverse GAN models in single image dehazing through benchmark datasets.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.34 no.10B
• /
• pp.1070-1075
• /
• 2009

In this paper, we propose new depth estimation and intermediate view synthesis algorithms for three-dimensional video generation. In order to improve temporal consistency of the depth map sequence, we add a temporal weighting function to the conventional matching function when we compute the matching cost for estimating the depth information. In addition, we propose a boundary noise removal method in the view synthesis operation. after finding boundary noise areas using the depth map, we replace them with corresponding texture information from the other reference image. Experimental results showed that the proposed algorithm improved temporal consistency of the depth sequence and reduced flickering artifacts in the virtual view. It also improved visual quality of the synthesized virtual views by removing the boundary noise.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.12 no.5
• /
• pp.467-472
• /
• 2002

In this paper, we propose a Fuzzy Cellular Neural Network(FCNN) that is based on a-Least Trimmed Square Hausdorff distance(a-LTSHD) which applies Hausdorff distance(HD) to the FCNN structure in order to remove the impulse noise of images effectively and also improve the speed of operation. FCNN incorporates Fuzzy set theory to Cellular Neural Network(CNN) structure and HD is used as a scale which computes the distance between set or two pixels in binary images without confrontation of the feature object. This method has been widely used with the adjustment of the object. For performance evaluation, our proposed method is analyzed in comparison with the conventional FCNN, with the Opening-Closing(OC) method, and the LTSHD based FCNN by using Mean Square Error(MSE) and Signal to Noise Ratio(SNR). As a result, the performance of our proposed network structure is found to be superior to the other algorithms in the removal of impulse noise.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.12 no.10
• /
• pp.1787-1793
• /
• 2008

Fingerprint Recognition System is made up of Off-line treatment and On-line treatment; the one is registering all the information of there trieving features which are retrieved in the digitalized fingerprint getting out of the analog fingerprint through the fingerprint acquisition device and the other is the treatment making the decision whether the users are approved to be accessed to the system or not with matching them with the fingerprint features which are retrieved and database from the input fingerprint when the users are approaching the system to use. In this paper, we propose a new data structure, called Union and Division, for processing binarized digital fingerprint image efficiently. We present a minutiae extraction algorithm that is using Union and Division and consists of binarization, noise removal, minutiae extraction stages.

• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• v.59 no.2
• /
• pp.145-154
• /
• 2023

In this study, to control the heading angle of a ship, which is constantly subjected to various internal and external disturbances during the voyage, an LADRC (linear active disturbance rejection control) design that focuses more on improving the disturbance removal performance was proposed. The speed rate of change of the ship's heading angle due to the turn of the rudder angle was selected as a significant factor, and the nonlinear model of the ship's maneuvering equation, including the steering gear, was treated as a total disturbance. It is the similar process with an LADRC design for the first-order transfer function model. At this time, the gains of the controller included in LADRC and the gains of the extended state observer were tuned to RCGAs (real-coded genetic algorithms) to minimize the integral time-weighted absolute error as an evaluation function. The simulation was performed by applying the proposed GA-LADRC controller to the heading angle control of the Mariner class vessel. In particular, it was confirmed that the proposed controller satisfactorily maintains and follows the set course even when the disturbances such as nonlinearity, modelling error, uncertainty and noise of the measurement sensor are considered.

• Journal of Korean Society of Transportation
• /
• v.26 no.4
• /
• pp.241-250
• /
• 2008

With the ever-growing amount of traffic, there is an increasing need for good quality travel time information. Various existing outlier filtering and missing data imputation algorithms using AVI data for interrupted and uninterrupted traffic flow have been proposed. This paper is devoted to development of an outlier filtering and missing data imputation algorithm by using Toll Collection System (TCS) data. TCS travel time data collected from August to September 2007 were employed. Travel time data from TCS are made out of records of every passing vehicle; these data have potential for providing real-time travel time information. However, the authors found that as the distance between entry tollgates and exit tollgates increases, the variance of travel time also increases. Also, time gaps appeared in the case of long distances between tollgates. Finally, the authors propose a new method for making representative values after removal of abnormal and "noise" data and after analyzing existing methods. The proposed algorithm is effective.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.12 no.10
• /
• pp.1897-1902
• /
• 2008

In computer vision analysis of detecting concrete slab surface cracks, there are many difficulties to overcome. Target images often have defamations due to the light condition and other external environment. Another difficulties in detecting concrete crack image is that there is no clear distinction in intensity between the crack and the surface since the surface is often irregular. In this paper, we apply ART2 based quantization in order to classify target concrete slab surface images into several areas with respect to the light intensity. From those quantized areas, we investigate the distribution of real cracks and noises. Then, we extract candidate crack areas after applying noise removal process to areas which have be th oracle and noises. Finally, crack areas are recognized by using morphological features of cracks from such candidate areas. In experiment with real world concrete slab structure images, our algorithm has advantage in recognizing accuracy of cracks to other algorithms especially in relatively brighter areas of concrete surface.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2009.10a
• /
• pp.530-533
• /
• 2009

This paper proposes a method of recognizing that a person doesn't put on a safety cap using image processing method in DSP(Digital Signal Processor). It processes inputted images by image input devices that equipped in a industrial settings. If the method recognizes a person that doesn't put on a safety cap, a system transfers relevant recognition result to a supervisor and takes proper measures. If an accident happens and someone doesn't put on a safety cap, additional casualities could be. Proposed method can nip additional casualties in the bud. To recognize that a person don't put on a safety cap, images are processed by object abstraction, removal of noise, decision of a thing or a person, abstraction of a head part in a image, recognizing whether a man puts on a safety cap using HSV color space or not, and so on. Image input and image process are processed by DSP. And C language-based codes are optimized by an eignefunction(Intrinsics) for speed improvement of algorithms.

• Journal of Information Processing Systems
• /
• v.14 no.6
• /
• pp.1445-1456
• /
• 2018

Environment perception and three-dimensional (3D) reconstruction tasks are used to provide unmanned ground vehicle (UGV) with driving awareness interfaces. The speed of obstacle segmentation and surrounding terrain reconstruction crucially influences decision making in UGVs. To increase the processing speed of environment information analysis, we develop a CPU-GPU hybrid system of automatic environment perception and 3D terrain reconstruction based on the integration of multiple sensors. The system consists of three functional modules, namely, multi-sensor data collection and pre-processing, environment perception, and 3D reconstruction. To integrate individual datasets collected from different sensors, the pre-processing function registers the sensed LiDAR (light detection and ranging) point clouds, video sequences, and motion information into a global terrain model after filtering redundant and noise data according to the redundancy removal principle. In the environment perception module, the registered discrete points are clustered into ground surface and individual objects by using a ground segmentation method and a connected component labeling algorithm. The estimated ground surface and non-ground objects indicate the terrain to be traversed and obstacles in the environment, thus creating driving awareness. The 3D reconstruction module calibrates the projection matrix between the mounted LiDAR and cameras to map the local point clouds onto the captured video images. Texture meshes and color particle models are used to reconstruct the ground surface and objects of the 3D terrain model, respectively. To accelerate the proposed system, we apply the GPU parallel computation method to implement the applied computer graphics and image processing algorithms in parallel.