http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Batkhuu Byambajav(바트후 뱜바자브),Jumabek Alikhanov(주마벡 알리하노브),Yang Fang(팡양),Seunghyun Ko(고승현),Geun Sik Jo(조근식) 한국지능정보시스템학회 2018 지능정보연구 Vol.24 No.1
Convolutional Neural Network (ConvNet) is one class of the powerful Deep Neural Network that can analyze and learn hierarchies of visual features. Originally, first neural network (Neocognitron) was introduced in the 80s. At that time, the neural network was not broadly used in both industry and academic field by cause of large-scale dataset shortage and low computational power. However, after a few decades later in 2012, Krizhevsky made a breakthrough on ILSVRC-12 visual recognition competition using Convolutional Neural Network. That breakthrough revived people interest in the neural network. The success of Convolutional Neural Network is achieved with two main factors. First of them is the emergence of advanced hardware (GPUs) for sufficient parallel computation. Second is the availability of large-scale datasets such as ImageNet (ILSVRC) dataset for training. Unfortunately, many new domains are bottlenecked by these factors. For most domains, it is difficult and requires lots of effort to gather large-scale dataset to train a ConvNet. Moreover, even if we have a large-scale dataset, training ConvNet from scratch is required expensive resource and time-consuming. These two obstacles can be solved by using transfer learning. Transfer learning is a method for transferring the knowledge from a source domain to new domain. There are two major Transfer learning cases. First one is ConvNet as fixed feature extractor, and the second one is Fine-tune the ConvNet on a new dataset. In the first case, using pre-trained ConvNet (such as on ImageNet) to compute feed-forward activations of the image into the ConvNet and extract activation features from specific layers. In the second case, replacing and retraining the ConvNet classifier on the new dataset, then fine-tune the weights of the pre-trained network with the backpropagation. In this paper, we focus on using multiple ConvNet layers as a fixed feature extractor only. However, applying features with high dimensional complexity that is directly extracted from multiple ConvNet layers is still a challenging problem. We observe that features extracted from multiple ConvNet layers address the different characteristics of the image which means better representation could be obtained by finding the optimal combination of multiple ConvNet layers. Based on that observation, we propose to employ multiple ConvNet layer representations for transfer learning instead of a single ConvNet layer representation. Overall, our primary pipeline has three steps. Firstly, images from target task are given as input to ConvNet, then that image will be feed-forwarded into pre-trained AlexNet, and the activation features from three fully connected convolutional layers are extracted. Secondly, activation features of three ConvNet layers are concatenated to obtain multiple ConvNet layers representation because it will gain more information about an image. When three fully connected layer features concatenated, the occurring image representation would have 9192 (4096+4096+1000) dimension features. However, features extracted from multiple ConvNet layers are redundant and noisy since they are extracted from the same ConvNet. Thus, a third step, we will use Principal Component Analysis (PCA) to select salient features before the training phase. When salient features are obtained, the classifier can classify image more accurately, and the performance of transfer learning can be improved. To evaluate proposed method, experiments are conducted in three standard datasets (Caltech-256, VOC07, and SUN397) to compare multiple ConvNet layer representations against single ConvNet layer representation by using PCA for feature selection and dimension reduction. Our experiments demonstrated the importance of feature selection for multiple ConvNet layer representation. Moreover, our proposed approach achieved 75.6% accuracy compared to 73.9% accuracy achieved by FC7 layer on the Caltech-256 dataset, 73.1% accuracy compared to
Caltech 보행자 감지를위한 Scale-aware Faster R-CNN
바트후 ( Batkhuu Byambajav ),주마벡 ( Jumabek Alikhanov ),조근식 ( Geun-sik Jo ) 한국정보처리학회 2016 한국정보처리학회 학술대회논문집 Vol.23 No.2
We present real-time pedestrian detection that exploit accuracy of Faster R-CNN network. Faster R-CNN has shown to success at PASCAL VOC multi-object detection tasks, and their ability to operate on raw pixel input without the need to design special features is very engaging. Therefore, in this work we apply and adjust Faster R- CNN to single object detection, which is pedestrian detection. The drawback of Faster R-CNN is its failure when object size is small. Previously, small sized object problem was solved by Scale-aware Network. We incorporate Scale-aware Network to Faster R-CNN. This made our method Scale-aware Faster R-CNN (DF R-CNN) that is both fast and very accurate. We separated Faster R-CNN networks into two sub-network,that is one for large-size objects and another one for small-size objects. The resulting approach achieves a 28.3% average miss rate on the Caltech Pedestrian detection benchmark, which is competitive with the other best reported results.
State of Forests and Biodiversity Conservation in Primorsky Krai, Russian Far East
Yuri Ivanovich Man'ko,이돈구,강호상,Batkhuu Nyam-Osor 한국산림과학회 2004 한국산림과학회지 Vol.93 No.6
The vegetation and species composition of Primorsky Krai located in Russian Far East are very similar to those in the Korean Peninsula and Northeastern part of China. The forests in Zapovednik(a strictly protected federal nature reserve) are unique old-growth forests without human disturbances for more than 500 years. The objectives of this study were to identify the forest resources and to suggest strategies for conservation of biodiversity and sustainable forest ecosystem management in Primorsky Krai. The total forestland comprises 11,833.3 thousand ha and is classified into 3 botanical-geographical zones; coniferous forest, coniferous-broadleaved mixed forest and forest-steppe. The total stock volume is estimated at 1,752 million m3, of which 66% are made up of conifers such as Picea jezoensis, Abies nephrolepis, Pinus koraiensis and Larix species. The flora contains 2,589 vascular plants and the forest plays important roles in the distribution and conservation of wildlife. The strategies for biodiversity conservation are as follows: 1) To conserve endemic and rare species of plant and wildlife with special protection, 2) To preserve the unique and original forest ecosystem without any industrial and human activities, and 3) To develop the sustainable uses and management of forest resources. The cooperative researches among Northeast Asian countries shall provide more detailed information not only on species distribution but on its biological and ecological characteristics.
Forest vegetation structure of Bogd Khan Mountain: A Strictly Protected Area in Mongolia
Badamtsetseg BAZARRAGCHAA,김현숙,Gantuya BATDELGER,Munkhjin Batkhuu,이상명,Seungah Yang,백운기,이중구 국립중앙과학관 2022 Journal of Asia-Pacific Biodiversity Vol.15 No.2
Bogd Khan Mountain is a strictly protected area located in the Khentei mountain range. It lies in the transitional zone of the Siberian coniferous taiga and the Asian steppe and has a unique ecosystem. The present study was conducted with the objective of identifying the plant community type, forest structure, and the changes in forest stands of the Bogd Khan Mountain. Sampling was done at 155 plots randomly along the gradients of the entire forest. Differences in species composition, biological spectrum, species diversity, and importance value were analyzed in each community. This study revealed the occurrence of six plant community types comparing of Larix sibirica, Picea obovata, Pinus sibirica, Betula pendula subsp. mandshurica, Pinus sylvestris and Populus tremula. These communities were different in species richness, diversity, and their distribution correlated to the altitudinal gradient (score 0.71735; p value 0.001). An upsurge in Picea obovata, and Betula pendula subsp. mandshurica community was found, indicating changes in the ecosystem such as permafrost melting, caused by anthropogenic influences. We suggest preserving the main dominant tree species in the native communities and reduction of the anthropogenic impacts urgently for the effective management of the biodiversity of the Bogd Khan Mountain.