• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.22 no.5
• /
• pp.7-14
• /
• 2021

Tomato crops are highly affected by tomato diseases, and if not prevented, a disease can cause severe losses for the agricultural economy. Therefore, there is a need for a system that quickly and accurately diagnoses various tomato diseases. In this paper, we propose a system that classifies nine diseases as well as healthy tomato plants by applying various pretrained deep learning-based CNN models trained on an ImageNet dataset. The tomato leaf image dataset obtained from PlantVillage is provided as input to ResNet, Xception, and DenseNet, which have deep learning-based CNN architectures. The proposed models were constructed by adding a top-level classifier to the basic CNN model, and they were trained by applying a 5-fold cross-validation strategy. All three of the proposed models were trained in two stages: transfer learning (which freezes the layers of the basic CNN model and then trains only the top-level classifiers), and fine-tuned learning (which sets the learning rate to a very small number and trains after unfreezing basic CNN layers). SGD, RMSprop, and Adam were applied as optimization algorithms. The experimental results show that the DenseNet CNN model to which the RMSprop algorithm was applied output the best results, with 98.63% accuracy.

• Journal of Broadcast Engineering
• /
• v.25 no.2
• /
• pp.208-217
• /
• 2020

The classification of plant diseases by images captured by a camera sensor has been studied over past decades. A method that has gained much interest is to use image segmentation, from which statistical features are derived and analyzed by machine learning. Recently, deep learning has been adopted in this area. However, image segmentation is still a difficult task to achieve stable performance due to a variety of environmental variations. The end-to-end learning in neural network has a demerit that train images may be different from real images acquired in outdoor fields. To solve these problems, we propose superpixel-based disease classification method using end-to-end CNN (convolutional neural network) learning. Based on experiments performed on PlantVillage apple images, the classification accuracy is 98.29% and 92.43% for full-image and superpixel. As well, the multivariate F1-score is (0.98, 0.93). Therefore we validate that the method of using superpixel is comparable to that of full-image.

• Proceedings of the KSRS Conference
• /
• 2008.10a
• /
• pp.29-32
• /
• 2008

In this paper, we propose one unsupervised classification technique using the degree of polarization (DoP) and the co-polarized phase-difference (CPD) statistics, instead of the entropy and alpha. It is shown that the DoP is closely related to the entropy, and the CPD to the alpha. The DoP explains the feature how much the effect of multiple reflections is contained. Hence, the DoP could be used as an important factor for classifying classes. The CPD can also be computed from the measured Mueller matrix elements. For the smooth surface scattering, the CPD is about $0^{\circ}$, and for dihedral-type scattering, the CPD is about $180^{\circ}$. A DoP-CPD diagram with appropriate boundaries between six different classes is developed based on the SAR image. The classification results are compared with the existing Entropy-alpha diagram as well as the IPL-AirSAR polarimetric data. The technique may have capability to classify an SAR image into six major classes; a bare surface, a village, a crown-layer short vegetation canopy, a trunk-layer short vegetation canopy, a crown-layer forest, and a trunk-dominated forest. Based on the DoP and CPD analysis, a simple three-component decomposition technique was also proposed.

• Journal of Korean Society of Rural Planning
• /
• v.28 no.3
• /
• pp.25-33
• /
• 2022

Due to the standardized urban development since the industrial era, many urban spaces have experienced non-placeness. Recently, although interest in urban landscape, history, and cultural elements has been increasing, landscape management and control measures are being implemented mainly in large cities due to the low status of rural areas in the country. From this point of view, this study tried to lay the foundation for basic research in related fields by classifying the characteristics and types of streetscapes located near large cities, and to suggest directions that should be considered when managing streetscapes in the future. As a result of the analysis of landscape adjectives, the village felt secluded due to the low density compared to the city street overall, but it did not have a unique image of the region. Three factors were derived through factor analysis, and preference was affected in the order of aesthetic, regularity, and uniqueness. In addition, the research site was classified into three types through cluster analysis, and it was confirmed that the differences by type were due to aesthetics and naturalness. On the other hand, the uniqueness is generally low in all regions, so it seems that fundamental countermeasures are needed.

• Korean Journal of Heritage: History & Science
• /
• v.53 no.2
• /
• pp.254-269
• /
• 2020

The purpose of this study is to propose guidance for wall renewal that is appropriate for an ancient city wall through application of advanced research and theories in wall design. It is a streetscape improvement project which forms part of the "Ancient City Image Finding Project". Study methods consist of advanced research classification, wall design theory contemplation, and analysis of the significance of designated ancient city areas and the "Ancient City Image Finding Project" status. Based on these methods, case study candidates were selected, case status and problems were identified, and improvement proposals were analyzed by comparing various features. Advanced wall research was classified into six categories including analysis of wall characteristics; wall design principle applications; wall structure, color, shape, and application; modern reinterpretation; palace walls; and house, temple, and village walls. The wall is an element of the streetscape improvement component of the "Ancient City Image Finding Project", with the characteristic of providing preceding experience in visual and cognitive awareness than interior structure. Case candidates for ancient city wall improvement are based on the composition distribution of the special conservation district in each ancient city as well as the conservation promotion district. Ultimately, the surrounding village of Gongju-si Geumseong-Dong Songsanri-gil, adjacent to the Royal Tomb of King Muryeong, was selected as the candidate. The "Ancient City Image Finding Project" of the surrounding village of Gongju-si Geumseong-Dong Songsanri-gil began with new Hanok construction. However, wall maintenance did not begin concurrently with that new Hanok construction. Support and maintenance took place afterwards as an exterior maintenance project for roadside structures. If the Hanok and wall were evaluated and constructed at the same time, the wall would have been built in unison with the size and design of the Hanok. The layout of the main building and wall of the Hanok is deemed to be a structure that is closed tightly because of its spatial proximity and tall height. Songsan-ri-gil's wall design should create a calm, subtle, and peaceful atmosphere with shapes, colors, and materials that express ancient city characteristics, but it is in an awkward position due to its sharpness and narrowness. The cause of the problem at Gongju-si Geumseong-dong Songsanri-gil, the case candidate, is that it is lacking significantly in terms of the aesthetic factors that traditional walls should possess. First, aesthetic consciousness seems to have disappeared during the selection and application process of the wall's natural materials. Second, the level of completion in design and harmony is absent. Maintenance guidance after analyzing the cause of problems in ancient city wall maintenance at Gongju-si Geumseong-dong Songsanri-gil, the subject area of research, is as follows: First, the Hanok design and layout of the wall and main gate should be reviewed simultaneously. Second, the one-sided use of natural stone wall in the Hanok wall design should be reexamined. Third, a permanent system to coordinate the opinions of citizens and experts during the planning and design phases should be employed. Fourth and finally, the Hanok's individuality shall be collectivized and its value as a cultural asset representing the identity of the community shall be increased.

• Korean Journal of Child Studies
• /
• v.28 no.3
• /
• pp.235-250
• /
• 2007

Using content analysis this study examined Korean people's traditional view on children with disabilities, hoping to contribute to the selection of appropriate picture books for an understanding of disabilities in kindergartens and child care centers. Story book characters with disabilities mainly interacted with their families and village people, had lower positions, did not accomplish their work on their own, and suffered from others' anger and beating. Differences in types of disabilities were that physical disabilities were viewed positively, while mental disabilities were viewed negatively. The former had wider scope of activities and maintained more positive relationships with other characters than the latter; furthermore, the former had higher position than the latter and had cooperative interactions with people without disabilities.

• Journal of the Korean housing association
• /
• v.15 no.4
• /
• pp.87-97
• /
• 2004

The purpose of this study is to introduce the design process for the 'natural-friendly' kindergarten playground using participatory design method. The 'A' kindergarten in Daejon was selected as a case study. The concept of the design was decided by the user's needs which draw out the pictures of children, owner's educational philosophy, and request of the teachers. As a result, the concept approached to 'our neighborhood (Korean traditional village)' image, and was planned with a focus on open space, which allows rich in natural elements(earth ground, trees, flowers vegetable farm, animals, water and sand etc.) and various play activities. And a user participatory design method proved that it was proper to design a kindergarten playground.

• Journal of Korean Society of Rural Planning
• /
• v.12 no.2 s.31
• /
• pp.1-9
• /
• 2006

Current environmental problems of rural area are connected to loss of rural functions which is food security as well as conservation of environment, balanced development of country and succession of traditional culture. To solve these problems, recent projects are focusing on social changes in rural area and conservation of rurality creating values of rural amenity resources. At this point, full implications of rurality which is the various aspects of rurality should be defined to be applied in direct plans to conserve the rurality. In this study, nine attributes of rurality are found as various meanings with adjectives included in image of rurality. For practical use of these adjectives of rurality attributes, we evaluate the list of rural amenity resources and suggested plans for conservation of rural amenity. These attributes can be used as an effective method for village plan which brings one of the attributes into relief.

• Proceeding of Spring/Autumn Annual Conference of KHA
• /
• 2003.11a
• /
• pp.153-157
• /
• 2003

The purpose of this study is to propose a ‘nature-friendly’ kindergarten playground design process, which is adopted from users needs. The concept of the design was decided by the pictures drawn by children, kindergarten owner's educational philosophy, and by the request of the teachers. As a result the concept approached to 'our neighborhood (Korean traditional village)'image, and was planned with a focus on an open space, which allows nature-like elements(earth ground, trees, flowers vegetable farm water and sand etc.) and various play activities.

• Journal of Korean Society of Rural Planning
• /
• v.23 no.1
• /
• pp.31-40
• /
• 2017

People are making efforts multilaterally to create agricultural landscape beautifully. Still in farming villages, however, large-scale facilities or houses are disharmonized with natural environment because of their outer colors, mostly primary colors, not considering natural environment. The main study is done on experts' color perception. Investigation about perception of color experts ' view according to the existence of the energy business based on the color to set the direction of the rural landscape in the future. The result of the study shows that the image of the rural area is not reflected well, and the symbolic color of the rural landscape is the reverse image according to the energy business. The rural village's image in the future, "Harmonious", "Natural", "Rural" were higher Figure. There were the most important and harmony with nature. Based on such future harmony with nature in the color planning The colors match and is expected to be the color within the scope. Lack of awareness in the current farming town scenery is unsatisfactory. Awareness about the need for further improvement was higher. Therefore, color according to a harmony with nature images in accordance with the type of farming model should be planned.