• Korean Journal of Soil Science and Fertilizer
• /
• v.49 no.6
• /
• pp.760-767
• /
• 2016

Food waste (FW) has been recognized as a critical problem in Korea and many research was conducted to efficiently utilize or treat FW. Main purpose of this research was to evaluate a feasibility for producing fermented organic fertilizer with vegetable waste (VW). Three different organic materials (saw dust, coco peat, and waste mushroom media) were mixed with VW at the rate of 30, 40, 50% respectively. Total days of composting experiment were 35 days and each sub samples were collected at every 5 days from starting of composting. Result showed that inner temperature of composting was increased to $60{\pm}4^{\circ}C$ within 5~10 days depending on varied organic materials and mixing ratio. Among different treatment, the highest increase of inner temperature was observed when 30% of saw dust was mixed with VW. After finishing composting experiment, maturity of each compost was evaluated with solvita and germination test. Maturity index (MI) of each treatment was ranged between 5~7 indicating that manufactured fertilizer was curing or finished stage. Calculated germination index (GI) was at the range of 57.83~101.16 depending on organic materials and mixing ratio. Both MI and GI showed that manufactured fertilizer was met for fertilizer criteria while control (VW only) was not adequate for composting. Overall, VW can be utilized for making organic fertilizer mixing with saw dust, coco peat and more research should be conducted to make high quality of organic fertilizer with vegetable waste.

• Korean Journal of Soil Science and Fertilizer
• /
• v.43 no.6
• /
• pp.858-863
• /
• 2010

The blueberry farming requires the soil condition of well-drainage, pH of 4.5 to 5.2, and high in organic matters for stable growth and development. Most of soil type of cultivated land in Korea, however, belongs to alkaline soils with low organic matter content and poor drainage. Therefore, the blueberry farmers use peat moss heavily to improve the soil condition, but the guideline on the effective and economic ratio of peat moss is not established yet. This study was performed to determine the cost effective peat moss ratio for amending soils, and to investigate the feasibility of using sawdust and coco peat as soil amendments. Peat moss, coco peat and sawdust are mixed with soil at the ratio of 0, 12.5, 50 and 100% (v/v). Among 3 organic materials with various mixture ratios, the pH of soil was the lowest in 100% peat moss and sawdust mixtures (pH 3.67 and pH 3.73, respectively), followed by pH 5.30 at 50% peat moss. The soil organic matter content are directly proportional to the mixture ratios in all three organic materials and the same trend was observed in the variation of content of exchangeable potassium in the coco peat treatments. On the contrary, the content of available phosphate, exchangeable calcium and magnesium decreased with increasing the ratio of organic materials. The nitrogen content in the leaves decreased as increasing the ratio of peat moss and coco peat in soil, but not of sawdust. The content of phosphate decreased but potassium increased as the ratio of sawdust and coco peat increased. There was no clear difference in the contents of magnesium and calcium among 3 organic materials. The plant height, stem diameter and dry weight of blueberry plants were the highest in 50 % peat moss, followed by 12.5% peat moss and 12.5% coco peat. The plants in 100% peat moss showed very poor growth. It can be concluded that peatmoss, when applied and managed appropriately, will be a good material for improving soil condition as well as securing desirable growth for blueberry. Upon coupling economic aspect, the optimum mixing ratio of peatmoss for blueberry farming is approximately 25-50%.

• Journal of the Korea Society of Computer and Information
• /
• v.28 no.9
• /
• pp.17-25
• /
• 2023

In this paper, we propose a model that can perform human pose estimation through a MobileViT-based model with fewer parameters and faster estimation. The based model demonstrates lightweight performance through a structure that combines features of convolutional neural networks with features of Vision Transformer. Transformer, which is a major mechanism in this study, has become more influential as its based models perform better than convolutional neural network-based models in the field of computer vision. Similarly, in the field of human pose estimation, Vision Transformer-based ViTPose maintains the best performance in all human pose estimation benchmarks such as COCO, OCHuman, and MPII. However, because Vision Transformer has a heavy model structure with a large number of parameters and requires a relatively large amount of computation, it costs users a lot to train the model. Accordingly, the based model overcame the insufficient Inductive Bias calculation problem, which requires a large amount of computation by Vision Transformer, with Local Representation through a convolutional neural network structure. Finally, the proposed model obtained a mean average precision of 0.694 on the MS COCO benchmark with 3.28 GFLOPs and 9.72 million parameters, which are 1/5 and 1/9 the number compared to ViTPose, respectively.

• Journal of Broadcast Engineering
• /
• v.27 no.1
• /
• pp.56-68
• /
• 2022

In this paper, we propose a compression method for multi-resolution feature maps for VCM. The proposed compression method removes the redundancy between the channels and resolution levels of the multi-resolution feature map through PCA-based transformation. According to each characteristic, the basis vectors and mean vector used for transformation, and the transformation coefficient obtained through the transformation are compressed using a VVC-based coder and DeepCABAC. In order to evaluate performance of the proposed method, the object detection performance was measured for the OpenImageV6 and COCO 2017 validation set, and the BD-rate of MPEG-VCM anchor and feature map compression anchor proposed in this paper was compared using bpp and mAP. As a result of the experiment, the proposed method shows a 25.71% BD-rate performance improvement compared to feature map compression anchor in OpenImageV6. Furthermore, for large objects of the COCO 2017 validation set, the BD-rate performance is improved by up to 43.72% compared to the MPEG-VCM anchor.

• International Journal of Advanced Culture Technology
• /
• v.10 no.2
• /
• pp.252-259
• /
• 2022

Artificial intelligence is used in fusion with image processing techniques using cameras. Image processing technology is a technology that processes objects in an image received from a camera in real time, and is used in various fields such as security monitoring and medical image analysis. If such image processing reduces the accuracy of recognition, providing incorrect information to medical image analysis, security monitoring, etc. may cause serious problems. Therefore, this paper uses a mixture of YOLOv4-tiny model and image processing algorithm and uses the COCO dataset for learning. The image processing algorithm performs five image processing methods such as normalization, Gaussian distribution, Otsu algorithm, equalization, and gradient operation. For RGB images, three image processing methods are performed: equalization, Gaussian blur, and gamma correction proceed. Among the nine algorithms applied in this paper, the Equalization and Gaussian Blur model showed the highest object detection accuracy of 96%, and the gamma correction (RGB environment) model showed the highest object detection rate of 89% outdoors (daytime). The image binarization model showed the highest object detection rate at 89% outdoors (night).

• IEMEK Journal of Embedded Systems and Applications
• /
• v.17 no.5
• /
• pp.273-280
• /
• 2022

Along with the advancement of deep learning technology, securing high-quality dataset for verification of developed technology is emerging as an important issue, and developing robust deep learning models to the domestic road environment is focused by many research groups. Especially, unlike expressways and automobile-only roads, in the complex city driving environment, various dynamic objects such as motorbikes, electric kickboards, large buses/truck, freight cars, pedestrians, and traffic lights are mixed in city road. In this paper, we built our dataset through multi camera-based processing (collection, refinement, and annotation) including the various objects in the city road and estimated quality and validity of our dataset by using YOLO-based model in object detection. Then, quantitative evaluation of our dataset is performed by comparing with the public dataset and qualitative evaluation of it is performed by comparing with experiment results using open platform. We generated our 2D dataset based on annotation rules of KITTI/COCO dataset, and compared the performance with the public dataset using the evaluation rules of KITTI/COCO dataset. As a result of comparison with public dataset, our dataset shows about 3 to 53% higher performance and thus the effectiveness of our dataset was validated.

• Journal of Animal Science and Technology
• /
• v.65 no.2
• /
• pp.377-386
• /
• 2023

The effect of different commercially available bedding materials on the growth performance and carcass characteristics of ducks for 42 days was investigated. 336 one-day-old White-Pekin ducklings (60.48 ± 0.16 g) were randomly allocated into 24-floor pens with one of the three beddings namely i) coco peat, ii) rice husks, or iii) sawdust. 14 ducklings per pen and 8 replicate pens per bedding material were used. Birds were fed a starter diet from days 1-21 and a grower diet from days 22-42. Weekly growth performance evaluation was conducted for the average body weight, weight gains, daily feed intake, and feed conversion efficiency. One bird per pen was sacrificed on day 42 for the evaluation of carcass characteristics including the carcass, breast, and leg muscle percentages. Breast and leg muscle samples were then collected and analyzed for their proximate and pH values. Higher body weights (p < 0.05) were noticed with rice husks on day 42 only. Improved daily gains (p < 0.05) were also noticed for birds raised with rice husks over the entire period (days 1-42). Concerning feed intake, higher values (p < 0.05) were similarly noted with rice husks for the grower phase (days 22-42), and the entire experimental period (days 1-42). Marginally improved feed intake values were also noted with the use of rice husks as the bedding materials on day 42 (p = 0.092). Improved feed efficiency (p < 0.05) was noticed with rice husks on day 35, the grower period, and the entire 42-day period. However, no significant differences were noticed for most of the carcass characteristics that were evaluated. Nevertheless, higher (p < 0.05) pH values for the breast muscle were noticed with the use of coco peat and sawdust as the bedding. Conclusively, the bedding type could have a significant impact on the growth performance of ducks without adverse effects on carcass characteristics. The use of rice husks as bedding might be advantageous and is therefore recommended.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2020.11a
• /
• pp.273-276
• /
• 2020

기존의 물리 센서를 활용한 주차 감지는 주차장 규모가 클수록 큰 비용이 필요하고 이미지 기반의 분석은 개별 주차장에 대한 데이터 라벨링과 학습의 노력이 필요했다. 본 논문은 LoRa(Long Range) 네트워크와 마이크로프로세서를 활용한 IoT기반의 시스템으로 영상데이터를 서버로 전송하고 COCO(Common Object in context) 데이터셋으로 학습된 Mask R-CNN 기반의 모델을 활용한 주차장 내 차량점유 감지 알고리즘을 통해 개별 주차장에 대한 학습 또는 라벨링 없이 주차장 내 주차상태를 식별하고 사용자에게 인터페이스를 통해 실시간으로 주차정보를 제공하는 시스템을 구현한다.

• KSBB Journal
• /
• v.11 no.4
• /
• pp.462-469
• /
• 1996

Biosolubilization of an Australian lignite was investigated by using Streptomyces viridosporus and Poria cocos. In order to solubilize coals effectively they were pretreated by nitric acid both in surface and liquid cultures. The optimum growth pH was 7.5 for S. viridosporus and 4.5 for P. cocos. The effects of various carbon, nitrogen and metal sources on overall solubilization were also studied. Solubility increased with the addition of urea for S. viridosporus, and peptone and tryptone for P. cocos. However carbon and metal sources had little or negative effects on solubilization. Maximum amount of coal solubilized was 85%(w/w) in a batch fermentation culture. Extracellular materials produced by micro-organism were found to be responsible for the coal solubilization. Approximately 70 to 80% of coal solubilization was determined to be the result of non-enzymatic reactions, and the rest to be the result of enzymatic reactions. Characteristics of the solubilized coal were compared with those of original coal and pretreated coal by the approximate and ultimate composition analysis, and IR-spectrum analysis. The spectroscopic results showed that the mechanism of coal solubilization was caused by continuous oxidation.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.38 no.3
• /
• pp.269-279
• /
• 2020

Traffic estimation mainly involves surveying equipment such as automatic vehicle classification, vehicle detection system, toll collection system, and personnel surveys through CCTV (Closed Circuit TeleVision), but this requires a lot of manpower and cost. In this study, we proposed a method of estimating traffic volume using deep learning and stereo CCTV to overcome the limitation of not detecting the entire vehicle in case of single CCTV. COCO (Common Objects in Context) dataset was used to train deep learning models to detect vehicles, and each vehicle was detected in left and right CCTV images in real time. Then, the vehicle that could not be detected from each image was additionally detected by using affine transformation to improve the accuracy of traffic volume. Experiments were conducted separately for the normal road environment and the case of weather conditions with fog. In the normal road environment, vehicle detection improved by 6.75% and 5.92% in left and right images, respectively, than in a single CCTV image. In addition, in the foggy road environment, vehicle detection was improved by 10.79% and 12.88% in the left and right images, respectively.