• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.38A no.3
• /
• pp.240-248
• /
• 2013

In this paper, we propose a super-resolution algorithm using an adaptive weighted interpolation(AWI) and discrete wavelet transform(DWT). In general, super-resolution algorithms for single-image, probability based operations have been used for searching high-frequency components. Consequently, the complexity of the algorithm is increased and it causes the increase of processing time. In the proposed algorithm, we first find high-frequency sub-bands by using DWT. Then we apply an AWI to the obtained high-frequency sub-bands to make them have the same size as the input image. Now, the interpolated high-frequency sub-bands and input image are properly combined and perform the inverse DWT. For the experiments, we use the down-sampled version of the original image($512{\times}512$) as a test image($256{\times}256$). Through experiment, we confirm the improved efficiency of the proposed algorithm comparing with interpolation algorithms and also save the processing time comparing with the probability based algorithms even with the similar performance.

• Journal of Korean Society of Transportation
• /
• v.25 no.2 s.95
• /
• pp.121-132
• /
• 2007

This paper addressed the issue of a generalized cost model for transit assignment. The model composed of walk time, waiting time (including transfer waiting time), line-haul time, transfer walk time, and fare. The weights of each component were supposed to be calculated using the stated preference (SP) data, which were collected prudently in order to reflect reality. The marginal rate of substitution and wage rate were applied to calculate the weights. The results showed that the weight of walking time per in-vehicle travel time (IVTT) was 1.507, the weight of waiting time (per IVTT) was 1.749, that of transfer time (per IVTT) was 1.474, and that of fare (per IVTT) was 1.476 for trips between inner-city areas in Seoul. Weights for each component were identified as 1.871, 1.967, 1.015, and 0.857, respectively, for trips between Seoul and Gyeonggi. Statistical significance existed between two cases and each variable was also statistically significant. Transit assignment using the relative weights estimated in this study was implemented to analyze the travel index in a macroscopic and quantitative basis. The results showed that average total travel times were 30.23 minutes and 63.29 minutes and average generalized costs were 2,510 won and 3,880 won for trips between inner-city areas in Seoul and between Seoul and Gyeonggi, respectively.

• Journal of KIISE:Computing Practices and Letters
• /
• v.12 no.5
• /
• pp.326-338
• /
• 2006

In this paper, we propose the weighted Markov model for recommending the users' prefered contents in the environment with considering the users' transition of their content consumption mind according to the kind of contents providing in time. In general, TV viewers have an intention to consume again the preferred contents consumed in recent by them. In order to take into the consideration, we modify the preference transition matrix by providing weights to the consecutively consumed contents for recommending the users' preferred contents. We applied the proposed model to the recommendation of TV viewer's genre preference. The experimental result shows that our method is more efficient than the typical methods.

• The KIPS Transactions:PartB
• /
• v.11B no.7 s.96
• /
• pp.849-854
• /
• 2004

This paper proposes an algorithm to enhance the execution time of kNN in the document classification. The proposed algorithm is to enhance the execution time by minimizing the computing cost of the similarity between two documents by using the list of pairs, while the conventional kNN uses the iist of pairs. The 1ist of pairs can be obtained by applying the matrix transposition to the list of pairs at the training phase of the document classification. This paper analyzed the proposed algorithm in the time complexity and compared it with the conventional kNN. And it compared the proposed algorithm with the conventional kNN by using routers-21578 data experimentally. The experimental results show that the proposed algorithm outperforms kNN about $90{\%}$ in terms of the ex-ecution time.

• The KIPS Transactions:PartD
• /
• v.17D no.6
• /
• pp.423-430
• /
• 2010

As software becomes large-scale and complicated, the need for Quality Assurance and management is increased and software testing is becoming more important. The main aims of software testing are not only detecting and handling the defects in the system but also investigating and managing the present system. But automatic testing tools require lots of time and efforts to detect and manage the risk in the system because test-cases used in the general automatic testing tools have the simply static information. In this thesis, the dynamic management technique for weighted testcases is designed to test the high-risk testcases preferentially by giving the testcases dynamic weight.

• Journal of Broadcast Engineering
• /
• v.26 no.6
• /
• pp.778-789
• /
• 2021

Recently, various lightweight methods for using Convolutional Neural Network(CNN) models in mobile devices have emerged. Weight quantization, which lowers bit precision of weights, is a lightweight method that enables a model to be used through integer calculation in a mobile environment where GPU acceleration is unable. Weight quantization has already been used in various models as a lightweight method to reduce computational complexity and model size with a small loss of accuracy. Considering the size of memory and computing speed as well as the storage size of the device and the limited network environment, this paper proposes a method of compressing integer weights after quantization using a video codec as a method. To verify the performance of the proposed method, experiments were conducted on VGG16, Resnet50, and Resnet18 models trained with ImageNet and Places365 datasets. As a result, loss of accuracy less than 2% and high compression efficiency were achieved in various models. In addition, as a result of comparison with similar compression methods, it was verified that the compression efficiency was more than doubled.

• The KIPS Transactions:PartB
• /
• v.8B no.6
• /
• pp.625-632
• /
• 2001

The electronic stores have realized that they need to understand their customers and to quickly response their wants and needs. To be successful in increasingly competitive Internet marketplace, recommender systems are adapting data mining techniques. One of most successful recommender technologies is collaborative filtering (CF) algorithm which recommends products to a target customer based on the information of other customers and employ statistical techniques to find a set of customers known as neighbors. However, the application of the systems, however, is not very suitable for seasonal products which are sensitive to time or season such as refrigerator or seasonal clothes. In this paper, we propose a new adjusted item-based recommendation generation algorithms called the exponentially weighted collaborative filtering recommendation (EWCFR) one that computes item-item similarities regarding seasonal products. Finally, we suggest the recommendation system with relatively high quality computing time on main memory database (MMDB) in XML since the collaborative filtering systems are needed that can quickly produce high quality recommendations with very large-scale problems.

• Journal of Korea Multimedia Society
• /
• v.11 no.4
• /
• pp.471-480
• /
• 2008

The deaf can not recognize useful sound informations such as alarm, doorbell, siren, car horn, and phone ring etc., because they have the hearing impairment. To solve this problems, portable hearing assistive devices which have suitable environment sound recognition methods are needed. In this paper, the DTW algorithm for sound recognition system with new active slope weighting constraint method was proposed. The environment sound recognition methods consist of three processes. First process is extraction of start point and end point using frequency and amplitude of sound. Second process is extraction of features and third process is classification of features for given segments. As a result of the experiment, the recognition rate of the proposed method is over 90%. And, the recognition rate of the proposed method increased about 20% than the conventional algorithm. Therefore if there are developed portable assistive devices which use developed method to recognize environment sound for hearing-impaired persons, they could be more convenient in life.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.17 no.3
• /
• pp.515-520
• /
• 2022

In context awareness using multiple sensors, when using sensor data detected and sent by each sensor, it is necessary to give different weights for each sensor. Even if the same type of sensor is configured for the same situation, sometimes it is necessary to assign different weights due to other secondary factors. It is inevitable to assign weights to events in the real world, and it can be said that a weighting method that can be used in a context awareness system using multiple sensors is necessary. In this study, we propose a weighting method for each sensor that reports to the host while the sensors continue to detect over time. In most IoT environments, the sensor continues the detection activity, and when the detected value shows a change pattern beyond a predetermined range, it is basically reported to the host. This can be called a kind of data stream environment. A weighting method was proposed for sensing data from multiple sensors in a data stream environment, and the new weighting method was to select and assign weights to data that indicates a context change in the stream.

• Journal of Korean Society of Transportation
• /
• v.33 no.5
• /
• pp.497-507
• /
• 2015

Nonrecurrent congestion of freeway was primarily caused by incident. The main cause of incident was known as a traffic accident. Therefore, accurate prediction of traffic incident clearance time is very important in accident management. Traffic accident data on freeway during year 2008 to year 2014 period were analyzed for this study. KNN(K-Nearest Neighbor) algorithm was hired for developing incident clearance time prediction model with the historical traffic accident data. Analysis result of accident data explains the level of accident significantly affect on the incident clearance time. For this reason, incident clearance time was categorized by accident level. Data were sorted by classification of traffic volume, number of lanes and time periods to consider traffic conditions and roadway geometry. Factors affecting incident clearance time were analyzed from the extracted data for identifying similar types of accident. Lastly, weight of detail factors was calculated in order to measure distance metric. Weight was calculated with applying standard method of normal distribution, then incident clearance time was predicted. Prediction result of model showed a lower prediction error(MAPE) than models of previous studies. The improve model developed in this study is expected to contribute to the efficient highway operation management when incident occurs.