• International Journal of Computer Science & Network Security
• /
• 제23권4호
• /
• pp.48-54
• /
• 2023

Nowadays, a blind man finds it very difficult to cross the roads. They should be very vigilant with every step they take. To resolve this problem, Convolutional Neural Networks(CNN) is a best method to analyse the data and automate the model without intervention of human being. In this work, a traffic signal recognition system is designed using CNN for the visually impaired. To provide a safe walking environment, a voice message is given according to light state and timer state at that instance. The developed model consists of two phases, in the first phase the CNN model is trained to classify different images captured from traffic signals. Common Objects in Context (COCO) labelled dataset is used, which includes images of different classes like traffic lights, bicycles, cars etc. The traffic light object will be detected using this labelled dataset with help of object detection model. The CNN model detects the color of the traffic light and timer displayed on the traffic image. In the second phase, from the detected color of the light and timer value a text message is generated and sent to the text-to-speech conversion model to make voice guidance for the blind person. The developed traffic light recognition model recognizes traffic light color and countdown timer displayed on the signal for safe signal crossing. The countdown timer displayed on the signal was not considered in existing models which is very useful. The proposed model has given accurate results in different scenarios when compared to other models.

• 한국방송∙미디어공학회:학술대회논문집
• /
• 한국방송∙미디어공학회 2022년도 하계학술대회
• /
• pp.140-142
• /
• 2022

최근 신경망 기반 기술들의 발달에 따라, 신경망 기술들은 충분히 높은 임무 수행 성능을 달성하고 있으며 사물인터넷, 스마트시티, 자율주행 등 다양한 환경을 고려한 응용 역시 활발히 연구되고 있다. 하지만 이러한 신경망의 임무 다양성과 복잡성은 더욱 많은 비디오 데이터가 요구되며 대역폭이 제한된 환경을 고려한 응용에서 이러한 비디오 데이터를 효과적으로 전송할 방법이 필요하다. 이에 따라 국제 표준화 단체인 MPEG 에서는 신경망 기계 소비에 적합한 비디오 부호화 표준 개발을 위해 Video Coding for Machines (VCM) 표준화를 진행하고 있다. 본 논문에서는 신경망의 특징 부호화 효율을 개선하기 위하여 VCM 을 위한 다중 스케일 특징 압축 방법을 제안한다. COCO2017 데이터셋의 검증 영상을 기반으로 제안방법을 평가한 결과, 압축된 특징의 크기는 원본 이미지의 0.03 배이며 6.8% 미만의 임무 정확도 손실을 보였다.

• 한국컴퓨터정보학회:학술대회논문집
• /
• 한국컴퓨터정보학회 2024년도 제69차 동계학술대회논문집 32권1호
• /
• pp.81-83
• /
• 2024

본 논문에서는 mask R-CNN의 이미지 세그먼테이션(Image Segmentation) 기법을 이용하여 철도의 선로를 식별하고 분류하는 방법을 제안한다. mask R-CNN의 이미지 세그먼테이션은 바운딩 박스(Bounding Box)를 통해 이미지에서 객체를 식별하는 R-CNN 알고리즘과는 달리 픽셀 단위로 관심 있는 객체를 검출하고 분류하는 기법으로서 오브젝트 디텍션(Object Detection)보다 더욱 정교한 객체 식별이 가능하다. 본 연구에서는 Pascal VOC 형태의 고속철도 데이터 24,205셋의 데이터를 전처리하고 MS COCO 데이터셋으로 변환하여, MMDetection의 mask R-CNN을 통해 픽셀 단위로 철도선로를 식별하고 정상/불량 상태를 분류하는 연구를 수행하였다. 선행연구에서는 YOLO를 활용하여 Polygon형태의 좌표를 바운딩 박스로 분류하였는데, 본 연구에서는 mask R-CNN을 활용함으로써 철도 선로를 더욱 정교하게 식별하였으며 정상/불량의 상태 분류는 YOLO와 유사한 성능을 보였다.

• 인터넷정보학회논문지
• /
• 제21권3호
• /
• pp.113-121
• /
• 2020

Human Pose Estimation (HPE) which localizes the human body joints becomes a high potential for high-level applications in the field of computer vision. The main challenges of HPE in real-time are occlusion, illumination change and diversity of pose appearance. The single RGB image is fed into HPE framework in order to reduce the computation cost by using depth-independent device such as a common camera, webcam, or phone cam. However, HPE based on the single RGB is not able to solve the above challenges due to inherent characteristics of color or texture. On the other hand, depth information which is fed into HPE framework and detects the human body parts in 3D coordinates can be usefully used to solve the above challenges. However, the depth information-based HPE requires the depth-dependent device which has space constraint and is cost consuming. Especially, the result of depth information-based HPE is less reliable due to the requirement of pose initialization and less stabilization of frame tracking. Therefore, this paper proposes a new method of HPE which is robust in estimating self-occlusion. There are many human parts which can be occluded by other body parts. However, this paper focuses only on head self-occlusion. The new method is a combination of the RGB image-based HPE framework and the depth information-based HPE framework. We evaluated the performance of the proposed method by COCO Object Keypoint Similarity library. By taking an advantage of RGB image-based HPE method and depth information-based HPE method, our HPE method based on RGB-D achieved the mAP of 0.903 and mAR of 0.938. It proved that our method outperforms the RGB-based HPE and the depth-based HPE.

• 한국측량학회지
• /
• 제36권5호
• /
• pp.423-432
• /
• 2018

Surface water mapping has been widely used in various remote sensing applications. Water indices have been commonly used to distinguish water bodies from land; however, determining the optimal threshold and discriminating water bodies from similar objects such as shadows and snow is difficult. Deep learning algorithms have greatly advanced image segmentation and classification. In particular, FCN (Fully Convolutional Network) is state-of-the-art in per-pixel image segmentation and are used in most benchmarks such as PASCAL VOC2012 and Microsoft COCO (Common Objects in Context). However, these data sets are designed for daily scenarios and a few studies have conducted on applications of FCN using large scale remotely sensed data set. This paper aims to fine-tune the pre-trained FCN network using the CRMS (Coastwide Reference Monitoring System) data set for surface water mapping. The CRMS provides color infrared aerial photos and ground truth maps for the monitoring and restoration of wetlands in Louisiana, USA. To effectively learn the characteristics of surface water, we used pre-trained the DeepWaterMap network, which classifies water, land, snow, ice, clouds, and shadows using Landsat satellite images. Furthermore, the DeepWaterMap network was fine-tuned for the CRMS data set using two classes: water and land. The fine-tuned network finally classifies surface water without any additional learning process. The experimental results show that the proposed method enables high-quality surface mapping from CRMS data set and show the suitability of pre-trained FCN networks using remote sensing data for surface water mapping.

• 한국약용작물학회지
• /
• 제19권6호
• /
• pp.441-445
• /
• 2011

This research was conducted to investigate the influence of variouis organic substrates on growth and yield of organically grown ginseng seedlings in a shaded plastic house. In the investigation of optimal substrate, the eight substrate were formulated by adjusting blending rate of peatmoss, perlite, coir dust(coco peat), and vermiculite. Then, the changes in physico chemical properties of root substrates as well as their influences on the growth characteristics and yield were determined at six months after sowing. The elevation of the blending rate of peatmoss from 50% to 70% with decrease in the rate of inorganic component (mixture of perlite and vermiculite) from 50 to 30% resulted in the increase in container capacities and decrease in total porosities and air-filled capacities. The concentrations of $NH_4-N$, $P_2O_5$ and K increased as the incorporation rate of castor seed meal, phosphate ore, and langbenite, respectively, were elevated during the root medium formulations. The PPV-1 and PPV-4 substrates produced high stem length, stem diameter, shoot fresh weight, leaf area and root length among eight substrate. Root fresh weight was heaviest in PPV-4 compound nursery media. The results of this experiment will be utilized in the new substrate application for ginseng organic culture in shaded vinyl house.

• 한국전자파학회논문지
• /
• 제22권11호
• /
• pp.1133-1136
• /
• 2011

본 논문에서는 ISM(산업/과학/의료) 대역(2.4~2.5 GHz)에서 IEEE 802.11b/g 적용을 위한 스트립라인 구조의 코리니어 안테나를 제안하였으며, 이는 동축 코리니어 안테나와 전 방향 평면 마이크로스트립 안테나(OMA)의 단점을 보완한 것이다. 4층 PCB 구조를 이용하여 기존 안테나에 비해 개선된 특성과 제작상의 이점을 가질 수 있었다. 안테나 배열을 위해 안테나 소자들의 외부 도체와 내부 도체를 ${\lambda}$/2 길이로 서로 엇갈리게 연결하여 동위상을 갖는 구조를 가지며, 상대 유전율 4.4, 손실 탄젠트(loss tangent) 0.02인 FR4 epoxy 기판을 이용하여 제작하였다. 최대 이득은 약 4.93 dBi가 측정되었으며, 기존 OMA 구조에 비하여 이득은 약 0.33 dBi 개선된 결과를 얻었다.

• 한국해양바이오학회지
• /
• 제4권1호
• /
• pp.24-30
• /
• 2010

Species Kappaphycus alvarezii (Doty) Doty, Kappaphycus striatum (Schmitz.) Doty and Eucheuma denticulatum (N. L. Burman) Collins et Harvey, which was brought to Vietnam from Japan in 1993 and Coco island, Martan Sea, Cebu, Philippines in 2005 have been cultivated in the different coasts of South Central Vietnam. Their growth rates and physical properties of carrageenan, then, were analyzed. The obtained results showed that the growth rate of E. denticulatum and K. striatum strains is higher than those of K. alvarezii. Species of K. striatum could grow over wide range of temperature and tolerate more strongly to high temperature compared with K. alvarezii, but their content and gel strength of kappa-carrageenan were almost the same and high. For purpose of the Kappahycus cultivation farms with stable and high production all year round (especially in the seawaters of shallow, semi-closed Lagoons where the water movement is not good and with high temperature in the hot season), mixed cropping of K. alvarezii and K. striatum as seeds stock during different cropping seasons was established. Our results suggested that K. alvarezii and K. striatum could be grown in the cool season (from Oct. to next March) with the same and high content and gel strength of kappa - carrageenan, but in the hot season need to chose K. striatum for cultivation only (from Apr. to Sept.).

• 한국환경복원기술학회지
• /
• 제18권1호
• /
• pp.115-125
• /
• 2015

This study was carried out to analyze comparison of carbon storage and physicochemical properties of vegetation soil for extensive green rooftop established at Seoul National University in september 2013. For this study, 42 plots were made by 2 kinds of vegetation soil including A-type and B-type. A-type vegetation soil plots were made of 90% perlite and 10% humus and B-type vegetation soil plots were made of 60% perlite, 20% vermiculite, 10% coco peat and 10% humus. This study used 6 kinds of plants which are Aster koraiensis, Sedum takesimense, Zoysia japonica Steud, Euonymus japonica, Rhododendron indicum SWEET and Ligustrum obtusifolium. Field research was carried out in 11 months after planting. Physiochemical properties of B-type vegetation soil plots were better than A-type vegetation soil plots in every way and soil carbon content was also higher at B-type vegetation soil plots as well. B-type vegetation soil plots were maintained 10 to 20% higher soil water content than A-type vegetation soil plots of the study period. The species of herb which showed the best carbon storage was Zoysia japonica Steud at B-type vegetation soil plots. The species of shrub which showed the best carbon storage was Ligustrum obtusifolium at B-type vegetation soil plots. Plants generally showed better growth at B-type vegetation soil plots and B-type vegetation soil plots were higher than A-type vegetation soil plots in soil carbon stock.

• 한국균학회지
• /
• 제44권3호
• /
• pp.127-131
• /
• 2016

2015년 9월 한국 화순 지역의 딸기와 파프리카 온실 내 묘상에서 이례적으로 버섯이 발생하였으며, 흔히 채소 식물체와 함께 자랐다. 필자들은 온실 내 발생한 버섯의 발생양상과 발생 버섯이 채소 작물의 생육과 품질에 미치는 영향을 조사하였으며, 온실 내 묘상으로부터 발생한 버섯 시료를 채집하여 형태적 및 분자생물학적 특성에 의해 동정하였다. 딸기 온실 채집 버섯 시료는 Leucocoprinus birnbaumii와 L. ianthinus로 동정되었으며, 파프리카 온실 채집버섯 시료는 Gymnopilus lepidotus로 동정되었다. 동정된 3종의 버섯 중에서 L. ianthinus와 G. lepidotus는 한국에서 처음 발견되었다. 조사 결과, 발생 버섯은 묘상에 사용된 코코피트에서 비롯된 것으로 밝혀졌으며, 온실 내 채소작물의 생육과 품질에는 영향을 미치지 않은 것으로 나타났다.