• Korean Journal of Agricultural Science
• /
• v.50 no.3
• /
• pp.357-364
• /
• 2023

In this study, we propose a ripe tomato bunch localization method based on convolutional neural networks, to be applied in robotic harvesting systems. Tomato images were obtained from a smart greenhouse at the Rural Development Administration (RDA). The sample images for training were extracted based on tomato maturity and resized to 128 × 128 pixels for use in the classification model. The model was constructed based on four-layer convolutional neural networks, and the classes were determined based on stage of maturity, using a Softmax classifier. The localization of the ripe tomato bunch region was indicated on a class activation map. The class activation map could show the approximate location of the tomato bunch but tends to present a local part or a large part of the ripe tomato bunch region, which could lead to poor performance. Therefore, we suggest a recursive method to improve the performance of the model. The classification results indicated that the accuracy, precision, recall, and F1-score were 0.98, 0.87, 0.98, and 0.92, respectively. The localization performance was 0.52, estimated by the Intersection over Union (IoU), and through input recursion, the IoU was improved by 13%. Based on the results, the proposed localization of the ripe tomato bunch area can be incorporated in robotic harvesting systems to establish the optimal harvesting paths.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.58 no.5
• /
• pp.1025-1034
• /
• 2009

In steel making production line, steel slabs are given a unique identification number. This identification number, Slab management number(SMN), gives information about the use of the slab. Identification of SMN has been done by humans for several years, but this is expensive and not accurate and it has been a heavy burden on the workers. Consequently, to improve efficiency, automatic recognition system is desirable. Generally, a recognition system consists of text localization, text extraction, character segmentation, and character recognition. For exact SMN identification, all the stage of the recognition system must be successful. In particular, the text localization is great important stage and difficult to process. However, because of many text-like patterns in a complex background and high fuzziness between the slab and background, directly extracting text region is difficult to process. If the slab region including SMN can be detected precisely, text localization algorithm will be able to be developed on the more simple method and the processing time of the overall recognition system will be reduced. This paper describes about the slab region localization using SIFT(Scale Invariant Feature Transform) features in the image. First, SIFT algorithm is applied the captured background and slab image, then features of two images are matched by Nearest Neighbor(NN) algorithm. However, correct matching rate can be low when two images are matched. Thus, to remove incorrect match between the features of two images, geometric locations of the matched two feature points are used. Finally, search rectangle method is performed in correct matching features, and then the top boundary and side boundaries of the slab region are determined. For this processes, we can reduce search region for extraction of SMN from the slab image. Most cases, to extract text region, search region is heuristically fixed [1][2]. However, the proposed algorithm is more analytic than other algorithms, because the search region is not fixed and the slab region is searched in the whole image. Experimental results show that the proposed algorithm has a good performance.

• International Journal of Oral Biology
• /
• v.36 no.4
• /
• pp.195-201
• /
• 2011

Matrix metalloproteinases (MMPs) have been implicated in tissue development and re-modeling. Dynamic morphological changes of tooth germs reflect involvement of these enzymes during odontogenesis. The present study was performed to investigate expression and localization of MMP-2 and MMP-9, which have been known to have type IV collagenase activities, in rat tooth germs at different developmental stages. MMP-2 expression was increased gradually in the tooth germs from cap to crown staged germs at both transcription and translation levels. The localization of this molecule was detected in secretory ameloblasts and preameloblasts. The strong immunoreactivities were occasionally seen along the basement membrane between ameloblasts (or preameloblasts) and odontoblasts (preodontoblasts). However, weak reactivity was detected in odontoblasts and reduced enamel epithelium. The level of MMP-9 expression in the tooth germs was higher in cap stage than in crown staged germs at both transcription and translation levels. They were strongly expressed in both ameloblasts and odontoblasts. Even though reduced enamel epithelium after enamel formation and inner enamel epithelium at the cap stage exhibited weak reactivity, strong reactivity was detected in dental follicles and perifollicular tissues surrounding cap staged germs. These results suggested that MMP-2 may involve degradation of the basement membrane during hard tissue formation, whereas MMP-9 might be involved in remodeling of follicular tissues.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.17 no.3
• /
• pp.297-303
• /
• 2007

In this paper, the stroke-based text extraction algorithm for digital video is proposed. The proposed algorithm consists of four stages such as text detection, text localization, text segmentation and geometric verification. The text detection stage ascertains that a given frame in a video sequence contains text. This procedure is accomplished by morphological operations for the pixels with higher possibility of being stroke-based text, which is called as seed points. For the text localization stage, morphological operations for the edges including seed points ate adopted followed by horizontal and vortical projections. Text segmentation stage is to classify projected areas into text and background regions according to their intensity distribution. Finally, in the geometric verification stage, the segmented area are verified by using prior knowledge of video text characteristics.

• International Journal of Oral Biology
• /
• v.43 no.4
• /
• pp.177-183
• /
• 2018

The objective of this study was to examine the expression pattern of Kelch-like ECH-associated protein 1 (Keap1) in the maxillary $2^{nd}$ molar germs of rats. We used the maxillary $2^{nd}$ molar germs in rats' pup at postnatal day 3 (bell stage), 6 (crown formation stage) and 9 (root formation stage). The investigation on mRNA and protein levels were done using reverse transcription - polymerase chain reaction and western blot. Localization of Keap 1 in the maxillary $2^{nd}$ molar germs were revealed through immunofluorescence staining. Keap1 from the maxillary 2nd molar germs were mostly manifested on postnatal day 3 and dramatically decreased on postnatal day 6 and 9 at mRNA and protein levels, while amelogenin and ameloblastin increased during the development of maxillary 2nd molar germs. During immunofluorescence analysis, the strong immunoreactivity against Keap1 was detected in the apical side of ameloblasts at the presecretory and secretory stages. However, Keap1 expression was hardly observed in the ameloblasts at the maturation stage. These results shows that Keap1 is strongly expressed in the presecretory and secretory ameloblasts of amelogenesis, and suggest that Keap1 may be a crucial molecule for the regulatory mechanisms tasked with the formation of enamel layer.

• IEIE Transactions on Smart Processing and Computing
• /
• v.2 no.2
• /
• pp.45-56
• /
• 2013

The precise identification of vehicle positions, known as the vehicle localization problem, is an important requirement for building intelligent vehicle ad-hoc networks (VANETs). To solve this problem, two categories of solutions are proposed: stand-alone and data fusion approaches. Compared to stand-alone approaches, which use single information including the global positioning system (GPS) and sensor-based navigation systems with differential corrections, data fusion approaches analyze the position information of several vehicles from GPS and sensor-based navigation systems, etc. Therefore, data fusion approaches show high accuracy. With the position information on a set of vehicles in the preprocessing stage, data fusion approaches is used to estimate the precise vehicular location in the local map building stage. This paper proposes an efficient local map building scheme, which increases the accuracy of the estimated vehicle positions via V2V communications. Even under the low ratio of vehicles with communication modules on the road, the proposed local map building scheme showed high accuracy when estimating the vehicle positions. From the experimental results based on the parameters of the practical vehicular environments, the accuracy of the proposed localization system approached the single lane-level.

• Proceedings of the KIEE Conference
• /
• 2011.07a
• /
• pp.1768-1769
• /
• 2011

The nonlinear estimation and the pseudo-linear estimation are used to treat the target localization in sensor network which provides range difference of arrival (RDOA) measurements. It is known that the nonlinear estimation has sensitive problem for the initial estimate and the pseudo-linear estimation has a large estimation error. The QCLS method is the typical estimator of the methods for pseudo-linear estimation. However the estimate by using the QCLS method includes the estimation error because the first stage of two estimation processes of the QCLS method causes the biased estimation error. Therefore we propose a instrumental variables(IV) method for minimizing the estimation error of the first stage. The simulation shows that the performance of the proposed method is superior to the QCLS method.

• Journal of the Korean association of regional geographers
• /
• v.12 no.1
• /
• pp.31-43
• /
• 2006

The aim of this paper is to explore localization processes of the Chuncheon mime festival (hereafter, CMF). Its history and processes can be periodized as three stages: the formative stage ($1989{\sim}1993$), the developing stage ($1994{\sim}1997$) and the settling-down & diffusing stage ($1998{\sim}2004$). Throughout this successful processes, CMF has contributed to settle down the structure of a virtuous circle in the regional (re)vitalization on the economic and socio-cultural dimension since the mid 1990s. This results from (1) a successful mix among vision providers, practioners, central and local governments, and citizen, (2) a interplay of the ability of spatial production by the mime culture and the capacity of cultural production by spatial occupiers.

• KIPS Transactions on Software and Data Engineering
• /
• v.10 no.9
• /
• pp.349-358
• /
• 2021

Recently, a method of finding the attention area or localization area for an object of an image using CAM (Class Activation Map)[1] has been variously carried out as a study of WSOL (Weakly Supervised Object Localization). The attention area extraction from the object heat map using CAM has a disadvantage in that it cannot find the entire area of the object by focusing mainly on the part where the features are most concentrated in the object. To improve this, using CAM and Selective Search[6] together, we first expand the attention area in the heat map, and a Gaussian smoothing is applied to the extended area to generate retraining data. Finally we train the data to expand the attention area of the objects. The proposed method requires retraining only once, and the search time to find an localization area is greatly reduced since the selective search is not needed in this stage. Through the experiment, the attention area was expanded from the existing CAM heat maps, and in the calculation of IOU (Intersection of Union) with the ground truth for the bounding box of the expanded attention area, about 58% was improved compared to the existing CAM.

• Smart Structures and Systems
• /
• v.6 no.5_6
• /
• pp.561-578
• /
• 2010

This study proposes a multi-level damage localization strategy to achieve an effective damage detection system for civil infrastructure systems based on wireless sensors. The proposed system is designed for use of distributed computation in a wireless sensor network (WSN). Modal identification is achieved using the frequency-domain decomposition (FDD) method and the peak-picking technique. The ASH (angle-between-string-and-horizon) and AS (axial strain) flexibility-based methods are employed for identifying and localizing damage. Fundamentally, the multi-level damage localization strategy does not activate all of the sensor nodes in the network at once. Instead, relatively few sensors are used to perform coarse-grained damage localization; if damage is detected, only those sensors in the potentially damaged regions are incrementally added to the network to perform finer-grained damage localization. In this way, many nodes are able to remain asleep for part or all of the multi-level interrogations, and thus the total energy cost is reduced considerably. In addition, a novel distributed computing strategy is also proposed to reduce the energy consumed in a sensor node, which distributes modal identification and damage detection tasks across a WSN and only allows small amount of useful intermediate results to be transmitted wirelessly. Computations are first performed on each leaf node independently, and the aggregated information is transmitted to one cluster head in each cluster. A second stage of computations are performed on each cluster head, and the identified operational deflection shapes and natural frequencies are transmitted to the base station of the WSN. The damage indicators are extracted at the base station. The proposed strategy yields a WSN-based SHM system which can effectively and automatically identify and localize damage, and is efficient in energy usage. The proposed strategy is validated using two illustrative numerical simulations and experimental validation is performed using a cantilevered beam.