• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.11
• /
• pp.3493-3506
• /
• 2022

In the field of scene segmentation, the precise segmentation of object boundaries in sports movement scene images is a great challenge. The geometric information and spatial information of the image are very important, but in many models, they are usually easy to be lost, which has a big influence on the performance of the model. To alleviate this problem, a parallel dense dilated convolution merging Network (termed PDDCM-Net) was proposed. The proposed PDDCMNet consists of a feature extractor, parallel dilated convolutions, and dense dilated convolutions merged with different dilation rates. We utilize different combinations of dilated convolutions that expand the receptive field of the model with fewer parameters than other advanced methods. Importantly, PDDCM-Net fuses both low-level and high-level information, in effect alleviating the problem of accurately segmenting the edge of the object and positioning the object position accurately. Experimental results validate that the proposed PDDCM-Net achieves a great improvement compared to several representative models on the COCO-Stuff data set.

• Journal of KIISE:Databases
• /
• v.28 no.4
• /
• pp.702-714
• /
• 2001

As Internet technology evolves, there is growing need of Internet queries involving multiple information sources. Efficient processing of such queries necessitates the integrated summary data that compactly represents the data distribution of the entire database scattered over many information sources. This paper presents an efficient method of managing the integrated summary data based on the wavelet transform and addresses Internet query processing using the integrated summary data. The simplest method for creating the integrated summary data would be to summarize the integrated data sidtribution obtained by merging the data distributions in multiple information sources. However, this method suffers from the high cost of transmitting storing and merging a large amount of data distribution. To overcome the drawbacks, we propose a new wavelet transform based method that creates the integrated summary data by merging multiple summary data and effective method for optimizing Internet queries using it A wavelet transformed summary data is converted to satisfy conditions for merging. Moreover i the merging process is very simpe owing to the properties of the wavelet transform. we formally derive the upper bound of the error of the wavelet transformed intergrated summary data. Compared with the histogram-based integrated summary data the wavelet transformedintegrated summary data provesto be 1.6~5.5 time more accurate when used for selectivity estimation in experiments. In processing Internet top-N queries involving 56 information sources using the integrated summary data reduces the processing cost to 1/44 of the cost of not using it.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.56 no.7
• /
• pp.1177-1182
• /
• 2007

Recently, IEC61850 supports the standardized communication technique in both station bus and process bus, and presents the substation automation model. In this paper, we propose the required techniques to develop the Merging Unit (MU), which is one of the important data acquisition equipment in substation automation, under IEC61850 communication protocol. Especially, we also propose the precision time synchronization technique using GPS(Global Positioning System) for the MU. IEC61850-9-2 SV (Sampled Value) service is applied to the MU which is designed using microprocessor. In order to evaluate the performance of the proposed MU, the playback program, which can reconstruct the transmitted sampled value data on the basis of time information, is developed and the performance of the IEC61850 SV service for MU is verified experimentally.

• Proceedings of the IEEK Conference
• /
• 1998.06a
• /
• pp.661-664
• /
• 1998

Since the region segmentation at high resolution contains most of viable semantic object contours in an image, the bottom-up approach for image segmentation is appropriate for the application such as MPEG-4 which needs to preserve semantic object contours. However, the conventioal region merging methods, that follow the region segmentation, have poor performance in keeping low-contrast semantic object contours. In this paper, we propose an image segmentation algorithm based on classified region merging. The algorithm pre-segments an image with a large number of small regions, and also classifies it into several classes having similar gradient characteristics. Then regions only in the same class are merged according to the boundary weakness or statisticsal similarity. The simulation result shows that the proposed image segmentation preserves semantic object contours very well even with a small number of regions.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.31B no.4
• /
• pp.106-114
• /
• 1994

A new merging technique is adopted for combining rectangular parallelepipes produced by 2-D rectangular code into more intuitive 30D volume elements. Rectangular parallelepiped codes (RP codes) can be used in volume-based representation of a three-dimensional object. We proposed more regularity-conserving 2-D rectangular coding scheme to merge rectangular cells represented by RP codes in three-dimensional space. After being constructed from modified 2-D rectangular code, 3-D RP codes are merged in the two orthogonal directions using new merging algorithm. The shape of merged 3-D object reconstructed by proposed algorithm is shown to be much closer to the original object shape than that of conventional RP codes. The storage requirement of merged object can be also reduced.

• Phonetics and Speech Sciences
• /
• v.6 no.3
• /
• pp.155-164
• /
• 2014

This paper proposes a new method to determine the recognition units for large vocabulary continuous speech recognition (LVCSR) in Korean by applying unsupervised segmentation and merging. In the proposed method, a text sentence is segmented into morphemes and position information is added to morphemes. Then submorpheme units are obtained by splitting the morpheme units through the maximization of posterior probability terms. The posterior probability terms are computed from the morpheme frequency distribution, the morpheme length distribution, and the morpheme frequency-of-frequency distribution. Finally, the recognition units are obtained by sequentially merging the submorpheme pair with the highest frequency. Computer experiments are conducted using a Korean LVCSR with a 100k word vocabulary and a trigram language model obtained by a 300 million eojeol (word phrase) corpus. The proposed method is shown to reduce the out-of-vocabulary rate to 1.8% and reduce the syllable error rate relatively by 14.0%.

• Industrial Engineering and Management Systems
• /
• v.12 no.3
• /
• pp.274-280
• /
• 2013

This paper proposes a space merging approach to studying the queuing models with finite buffers and jump priorities. Upon the arrival of a call with low priority, one call of such kind is assumed to be transferred to the end of the queue of high priority calls. The transfer probabilities depend on the state of the queue of the heterogeneous calls. We developed the algorithms to calculate the quality of service metrics of such queuing models, and the results of the numerical experiments are shown.

• Proceedings of the IEEK Conference
• /
• 1998.10a
• /
• pp.903-906
• /
• 1998

We propose a new method for merging partical regions which have similar motion characteristics. In the proposed method, the trajectories of each partial regions are computed and the regions consisting a same object are merged by a similarity function defined in this paper. The similarity function is defined to take into consideration the motion characteristics of the trajectories. The proposed method shows good performance when the regions of an object show little variation in size and shape.

• Proceedings of the IEEK Conference
• /
• 2003.11b
• /
• pp.99-102
• /
• 2003

In this paper we propose an adaptive learning method for HMM classifiers by using splitting and merging techniques to overcome the problem of the conventional teaming, where one HMM classifier per class has been trained, individually. The experimental results demonstrate a possibility that the proposed mechanism could be applied for applications of having multiple clusters in a class.

• Transactions on Electrical and Electronic Materials
• /
• v.15 no.3
• /
• pp.125-129
• /
• 2014

A dataset can be clustered by merging the bucket indices that come from the random projection of locality sensitive hashing functions. It should be noted that for this to work the merging interval must be calculated first. To improve the feasibility of large scale data clustering in high dimensional space we propose an enhanced Locality Sensitive Hashing Clustering Method. Firstly, multiple hashing functions are generated. Secondly, data points are projected to bucket indices. Thirdly, bucket indices are clustered to get class labels. Experimental results showed that on synthetic datasets this method achieves high accuracy at much improved cluster speeds. These attributes make it well suited to clustering data in high dimensional space.