• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.1
• /
• pp.60-79
• /
• 2022

The U-Net architecture-based segmentation models attained remarkable performance in numerous medical image segmentation missions like skin lesion segmentation. Nevertheless, the resolution gradually decreases and the loss of spatial information increases with deeper network. The fusion of adjacent layers is not enough to make up for the lost spatial information, thus resulting in errors of segmentation boundary so as to decline the accuracy of segmentation. To tackle the issue, we propose a new deep learning-based segmentation model. In the decoding stage, the feature channels of each decoding unit are concatenated with all the feature channels of the upper coding unit. Which is done in order to ensure the segmentation effect by integrating spatial and semantic information, and promotes the robustness and generalization of our model by combining the atrous spatial pyramid pooling (ASPP) module and channel attention module (CAM). Extensive experiments on ISIC2016 and ISIC2017 common datasets proved that our model implements well and outperforms compared segmentation models for skin lesion segmentation.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.41 no.4
• /
• pp.9-21
• /
• 2004

The current paper proposes a genetic algorithm (GA)-based segmentation method that can automatically extract and track moving objects. The proposed method mainly consists of spatial and temporal segmentation; the spatial segmentation divides each frame into regions with accurate boundaries, and the temporal segmentation divides each frame into background and foreground areas. The spatial segmentation is performed using chromosomes that evolve distributed genetic algorithms (DGAs). However, unlike standard DGAs, the chromosomes are initiated from the segmentation result of the previous frame, then only unstable chromosomes corresponding to actual moving object parts are evolved by mating operators. For the temporal segmentation, adaptive thresholding is performed based on the intensity difference between two consecutive frames. The spatial and temporal segmentation results are then combined for object extraction, and tracking is performed using the natural correspondence established by the proposed spatial segmentation method. The main advantages of the proposed method are twofold: First, proposed video segmentation method does not require any a priori information second, the proposed GA-based segmentation method enhances the search efficiency and incorporates a tracking algorithm within its own architecture. These advantages were confirmed by experiments where the proposed method was success fully applied to well-known and natural video sequences.

• Korean Journal of Remote Sensing
• /
• v.26 no.6
• /
• pp.731-740
• /
• 2010

A region merging technique is suggested in this paper for the segmentation of high-spatial resolution imagery. It employs a region growing scheme based on the region adjacency graph (RAG). The proposed algorithm uses directional neighbor-line average feature vectors to improve the quality of segmentation. The feature vector consists of 9 components which includes an observation and 8 directional averages. Each directional average is the average of the pixel values along the neighbor line for a given neighbor line length at each direction. The merging coefficients of the segmentation process use a part of the feature components according to a given merging coefficient order. This study performed the extensive experiments using simulation data and a real high-spatial resolution data of IKONOS. The experimental results show that the new approach proposed in this study is quite effective to provide segments of high quality for the object-based analysis of high-spatial resolution images.

• Proceedings of the KSRS Conference
• /
• 2003.11a
• /
• pp.153-155
• /
• 2003

Laser scanning has become a viable technique for the collection of a large amount of accurate 3D point data densely distributed on the scanned object surface. The inherent 3D nature of the sub-randomly distributed point cloud provides abundant spatial information. To explore valuable spatial information from laser scanned data becomes an active research topic, for instance extracting digital elevation model, building models, and vegetation volumes. The sub-randomly distributed point cloud should be segmented and classified before the extraction of spatial information. This paper investigates some exist segmentation methods, and then proposes an octree-based split-and-merge segmentation method to divide lidar data into clusters belonging to 3D planes. Therefore, the classification of lidar data can be performed based on the derived attributes of extracted 3D planes. The test results of both ground and airborne lidar data show the potential of applying this method to extract spatial features from lidar data.

• Proceedings of the KSRS Conference
• /
• 2002.10a
• /
• pp.59-64
• /
• 2002

This paper presents a method for line segment extraction for 3-d building reconstruction. Building roofs are described as a set of planar polygonal patches, each of which is extracted by watershed-based image segmentation, line segment matching and coplanar grouping. Coplanar grouping and polygonal patch formation are performed per region by selecting 3-d line segments that are matched using epipolar geometry and flight information. The algorithm has been applied to high resolution aerial images and the results show accurate 3-d building reconstruction.

• Proceedings of the KSRS Conference
• /
• 2003.11a
• /
• pp.1323-1325
• /
• 2003

As to many conventional segmentation approaches , spatial autocorrelation, perhaps being the first law of geography, is always overlooked. Thus, the corresponding segmentation results are always not so satisfying, which will further affect the subsequent image processing or analyses. In order to improve segmentation results, a geostatistic based segmentation approach with the consideration of spatial autocorrelation hidden in remote-sensing images is proposed in this article. First, by calculating the mean variance between each pair of pixels at given different lag distances, information like the size of typical targets in the scene can be obtained, and segmentation thresholds are calculated accordingly. Second, an initial region growing segmentation approach is implemented. Finally, based on the segmentation thresholds obtained at the first step and the initial segmentation results, the final segmentation results are obtained using the same region growing approach by taking the local mutual best fitting strategy. From the experiment results, we found the approach is rather promising. However, there still exists some problems to be settled, and further researches should be conducted in the future.

• Journal of the Korean Regional Science Association
• /
• v.12 no.2
• /
• pp.37-57
• /
• 1996

Intra-metropolitan spatial segmentation of the labor marker requires barriers of mobility on both supply and demand side of the local labor marker. The phenomena of spatial segmentation of the labor market are particularly applied to the secondary workers rather than to the primary workers. Supply side barriers include the costs of obtaining job information regarding jobs outside of the immediate area, commuting costs, and barriers to residential mobility. Demand side barriers include site-specific technology and product demand, and discrimination. In this paper, I discuss these barriers and examine their implications for differences in segmentation by demographic and skill groups at the intra-metropolitan scale. In particular, I apply a job search model to examine supply side barriers such as information and commuting costs, and an implicit contract model to explain demand side barriers such as dual/internal labor market and firms' (re) location strategies.

• Journal of the Korean Geographical Society
• /
• v.48 no.6
• /
• pp.978-993
• /
• 2013

It can be advantage as well as disadvantage to use the spatial weight matrix in a spatial regression model; it would benefit from explicitly quantifying spatial relationships between geographical units, but necessarily involve subjective judgment while specifying the matrix. We took Incheon City as a study area and investigated how the fitness of a spatial regression model changed by constructing various spatial weight matrices. In addition, we explored neighborhood segmentation in the study area and analyzed any influence of it on the model adequacy of two basic spatial regression models, i.e., spatial lagged and spatial error models. The results showed that it can help to improve the adequacy of models to specify the spatial weight matrix strictly, that is, interpreting the neighborhood as small as possible when estimating land price. It was also found that the spatial error model would be preferred in the area with serious spatial heterogeneity. In such area, we found that its spatial heterogeneity can be alleviated by delineating sub-neighborhoods, and as a result, the spatial lagged model would be preferred over the spatial error model.

• Proceedings of the KSRS Conference
• /
• 2003.11a
• /
• pp.204-206
• /
• 2003

In order to utilize remote sensed images effectively, a lot of image classification methods are suggested for many years. But, the accuracy of traditional methods based on pixel-based classification is not high in general. In this study, object oriented classification based on image segmentation is used to classify Landsat images. A necessary prerequisite for object oriented image classification is successful image segmentation. Object oriented image classification, which is based on fuzzy logic, allows the integration of a broad spectrum of different object features, such as spectral values , shape and texture. Landsat images are divided into urban, agriculture, forest, grassland, wetland, barren and water in sochon-gun, Chungcheongnam-do using object oriented classification algorithms in this paper. Preliminary results will help to perform an automatic image classification in the future.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.6
• /
• pp.3121-3143
• /
• 2019

Segmentation plays an important role in the field of image processing and computer vision. Intuitionistic fuzzy C-means (IFCM) clustering algorithm emerged as an effective technique for image segmentation in recent years. However, standard fuzzy C-means (FCM) and IFCM algorithms are sensitive to noise and initial cluster centers, and they ignore the spatial relationship of pixels. In view of these shortcomings, an improved algorithm based on IFCM is proposed in this paper. Firstly, we propose a modified non-membership function to generate intuitionistic fuzzy set and a method of determining initial clustering centers based on grayscale features, they highlight the effect of uncertainty in intuitionistic fuzzy set and improve the robustness to noise. Secondly, an improved nonlinear kernel function is proposed to map data into kernel space to measure the distance between data and the cluster centers more accurately. Thirdly, the local spatial-gray information measure is introduced, which considers membership degree, gray features and spatial position information at the same time. Finally, we propose a new measure of intuitionistic fuzzy entropy, it takes into account fuzziness and intuition of intuitionistic fuzzy set. The experimental results show that compared with other IFCM based algorithms, the proposed algorithm has better segmentation and clustering performance.