• Journal of the Korea Institute of Building Construction
• /
• v.23 no.2
• /
• pp.197-207
• /
• 2023

Concrete is a widely used material due to its excellent compressive strength and durability. However, depending on the surrounding environment and the characteristics of the materials used in the construction, various defects may occur, such as cracks on the surface and subsidence of the structure. The detects on the surface of the concrete structure occur after completion or over time. Neglecting these cracks may lead to severe structural damage, necessitating regular safety inspections. Traditional visual inspections of concrete walls are labor-intensive and expensive. This research presents a deep learning-based semantic segmentation model designed to detect cracks in concrete walls. The model addresses surface defects that arise from aging, and an image augmentation technique is employed to enhance feature extraction and generalization performance. A dataset for semantic segmentation was created by combining publicly available and self-generated datasets, and notable semantic segmentation models were evaluated and tested. The model, specifically trained for concrete wall fracture detection, achieved an extraction performance of 81.4%. Moreover, a 3% performance improvement was observed when applying the developed augmentation technique.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• fall
• /
• pp.290-293
• /
• 2021

In this paper, we implement a learning efficiency verification system to inspire learning motivation and help improve concentration by detecting the situation of the user studying. To this aim, data on learning attitude and concentration are measured by extracting the movement of the user's face or body through a real-time camera. The Jetson board was used to implement the real-time embedded system, and a convolutional neural network (CNN) was implemented for image recognition. After detecting the feature part of the object using a CNN, motion detection is performed. The captured image is shown in a GUI written in PYQT5, and data is collected by sending push messages when each of the actions is obstructed. In addition, each function can be executed on the main screen made with the GUI, and functions such as a statistical graph that calculates the collected data, To do list, and white noise are performed. Through learning efficiency checking system, various functions including data collection and analysis of targets were provided to users.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2003.05a
• /
• pp.165-170
• /
• 2003

This paper presents an integrated approach to unconstrained face recognition in arbitrary scenes. The front end of the system comprises of a scale and pose tolerant face detector. Scale normalization is achieved through novel combination of a skin color segmentation and log-polar mapping procedure. Principal component analysis is used with the multi-view approach proposed in[10] to handle the pose variations. For a given color input image, the detector encloses a face in a complex scene within a circular boundary and indicates the position of the nose. Next, for recognition, a radial grid mapping centered on the nose yields a feature vector within the circular boundary. As the width of the color segmented region provides an estimated size for the face, the extracted feature vector is scale normalized by the estimated size. The feature vector is input to a trained neural network classifier for face identification. The system was evaluated using a database of 20 person's faces with varying scale and pose obtained on different complex backgrounds. The performance of the face recognizer was also quite good except for sensitivity to small scale face images. The integrated system achieved average recognition rates of 87% to 92%.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.7 no.6
• /
• pp.1312-1317
• /
• 2003

This paper presents an integrated approach to unconstrained face recognition in arbitrary scenes. The front end of the system comprises of a scale and pose tolerant face detector. Scale normalization is achieved through novel combination of a skin color segmentation and log-polar mapping procedure. Principal component analysis is used with the multi-view approach proposed in[10] to handle the pose variations. For a given color input image, the detector encloses a face in a complex scene within a circular boundary and indicates the position of the nose. Next, for recognition, a radial grid mapping centered on the nose yields a feature vector within the circular boundary. As the width of the color segmented region provides an estimated size for the face, the extracted feature vector is scale normalized by the estimated size. The feature vector is input to a trained neural network classifier for face identification. The system was evaluated using a database of 20 person's faces with varying scale and pose obtained on different complex backgrounds. The performance of the face recognizer was also quite good except for sensitivity to small scale face images. The integrated system achieved average recognition rates of 87% to 92%.

• Journal of KIISE:Computing Practices and Letters
• /
• v.16 no.6
• /
• pp.654-662
• /
• 2010

In this paper, we suggest an advanced algorithm, to recognize pedestrian/non-pedestrian using second-stage cascade method, which applies Adaboost algorithm to make a strong classification from weak classifications. First, we extract two feature vectors: (i) Histogram of Oriented Gradient (HOG) which includes gradient information and differential magnitude; (ii) Curvature-HOG which is based on four different curvature features per pixel. And then, a strong classification needs to be obtained from weak classifications for composite recognition method using both HOG and curvature-HOG. In the proposed method, we use one feature vector and one strong classification for the first stage of recognition. For the recognition-failed image, the other feature and strong classification will be used for the second stage of recognition. Based on our experiment, the proposed algorithm shows higher recognition rate compared to the traditional method.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.3
• /
• pp.701-720
• /
• 2023

Aerial objects are more challenging to segment than normal objects, which are usually smaller and have less textural detail. In the process of segmentation, target objects are easily omitted and misdetected, which is problematic. To alleviate these issues, we propose local aggregation feature pyramid networks (LAFPNs) and pyramid integrated context modules (PICMs) for aerial object segmentation. First, using an LAFPN, while strengthening the deep features, the extent to which low-level features interfere with high-level features is reduced, and numerous dense and small aerial targets are prevented from being mistakenly detected as a whole. Second, the PICM uses global information to guide local features, which enhances the network's comprehensive understanding of an entire image and reduces the missed detection of small aerial objects due to insufficient texture information. We evaluate our network with the MS COCO dataset using three categories: airplanes, birds, and kites. Compared with Mask R-CNN, our network achieves performance improvements of 1.7%, 4.9%, and 7.7% in terms of the AP metrics for the three categories. Without pretraining or any postprocessing, the segmentation performance of our network for aerial objects is superior to that of several recent methods based on classic algorithms.

• KIPS Transactions on Software and Data Engineering
• /
• v.2 no.10
• /
• pp.713-722
• /
• 2013

This paper proposes a method of using depth information to detect the hand region in real-time based on the cascade method. In order to ensure stable and speedy detection of the hand region even under conditions of lighting changes in the test environment, this study uses only features based on depth information, and proposes a method of detecting the hand region by means of a classifier that uses boosting and cascading methods. First, in order to extract features using only depth information, we calculate the difference between the depth value at the center of the input image and the average of depth value within the segmented block, and to ensure that hand regions of all sizes will be detected, we use the central depth value and the second order linear model to predict the size of the hand region. The cascade method is applied to implement training and recognition by extracting features from the hand region. The classifier proposed in this paper maintains accuracy and enhances speed by composing each stage into a single weak classifier and obtaining the threshold value that satisfies the detection rate while exhibiting the lowest error rate to perform over-fitting training. The trained classifier is used to classify the hand region, and detects the final hand region in the final merger stage. Lastly, to verify performance, we perform quantitative and qualitative comparative analyses with various conventional AdaBoost algorithms to confirm the efficiency of the hand region detection algorithm proposed in this paper.

• Journal of the Korea Convergence Society
• /
• v.10 no.5
• /
• pp.1-8
• /
• 2019

In this paper, we propose an improved facial skin color extraction method to solve the problem that facial surface is lost due to shadow or illumination in skin color extraction process and skin color extraction is not possible. In the conventional HSV method, when facial surface is brightly illuminated by light, the skin color component is lost in the skin color extraction process, so that a loss area appears on the face surface. In order to solve these problems, we extract the skin color, determine the elements in the H channel value range of the skin color in the HSV color space among the lost skin elements, and combine the coordinates of the lost part with the coordinates of the original image, To minimize the number of In the face detection process, the face was detected using the LBP Cascade Classifier, which represents texture feature information in the extracted skin color image. Experimental results show that the proposed method improves the detection rate and accuracy by 5.8% and 9.6%, respectively, compared with conventional RGB and HSV skin color extraction and face detection using the LBP cascade classifier method.

• Journal of the Korean Society of Radiology
• /
• v.18 no.5
• /
• pp.557-566
• /
• 2024

Ultrasound-guided regional anesthesia is one of the most common techniques used in peripheral nerve blockade by enhancing pain control and recovery time. However, accurate Brachial Plexus (BP) nerve detection and identification remains a challenging task due to the difficulty in data acquisition such as speckle and Doppler artifacts even for experienced anesthesiologists. To mitigate the issue, we introduce a BP nerve small target segmentation network by incorporating BP object detection and U-Net based semantic segmentation into a single deep learning framework based on the multi-scale approach. To this end, the current BP detection and identification was estimated: 1) A RetinaNet model was used to roughly locate the BP nerve region using multi-scale based feature representations, and 2) U-Net was then used by feeding plural BP nerve features for each scale. The experimental results demonstrate that our proposed model produces high quality BP segmentation by increasing the accuracies of the BP nerve identification with the assistance of roughly locating the BP nerve area compared to competing methods such as segmentation-only models.

• The KIPS Transactions:PartB
• /
• v.10B no.2
• /
• pp.135-142
• /
• 2003

Robust extraction of 3D object's features, shape and global motion information from 2D image sequence is described. The object's 21 feature points on the pyramid type synthetic object are extracted automatically using color transform technique. The extracted features are used to recover the 3D shape and global motion of the object using stereo paraperspective camera model and sequential SVD(Singuiar Value Decomposition) factorization method. An inherent error of depth recovery due to the paraperspective camera model was removed by using the stereo image analysis. A 30 synthetic object with 21 features reflecting various position was designed and tested to show the performance of proposed algorithm by comparing the recovered shape and motion data with the measured values.