• The KIPS Transactions:PartB
• /
• v.16B no.4
• /
• pp.299-308
• /
• 2009

AAM(Active Appearance Model) is an algorithm to extract face feature points with statistical models of shape and texture information based on PCA(Principal Component Analysis). This method is widely used for face recognition, face modeling and expression recognition. However, the detection performance of AAM algorithm is sensitive to initial value and the AAM method has the problem that detection error is increased when an input image is quite different from training data. Especially, the algorithm shows high accuracy in case of closed lips but the detection error is increased in case of opened lips and deformed lips according to the facial expression of user. To solve these problems, we propose the improved AAM algorithm using lip feature points which is extracted based on a new lip detection algorithm. In this paper, we select a searching region based on the face feature points which are detected by AAM algorithm. And lip corner points are extracted by using Canny edge detection and histogram projection method in the selected searching region. Then, lip region is accurately detected by combining color and edge information of lip in the searching region which is adjusted based on the position of the detected lip corners. Based on that, the accuracy and processing speed of lip detection are improved. Experimental results showed that the RMS(Root Mean Square) error of the proposed method was reduced as much as 4.21 pixels compared to that only using AAM algorithm.

• Korean Journal of Remote Sensing
• /
• v.31 no.5
• /
• pp.395-408
• /
• 2015

High-resolution satellite image mosaics are becoming increasingly important in the field of remote sensing image analysis as an essential image processing to create a large image constructed from several smaller images. In this paper, we present an automatic seamline extraction technique and the procedure to generate a mosaic image by this technique. For more effective seamline extraction in the overlap region of adjacent images, an NDVI-based seamline extraction technique is developed, which takes advantage of the computational time and memory. The Normalized Difference Vegetation Index(NDVI) is an index of plant "greeness" or photosynthetic activity that is employed to extract the initial seamline. The NDVI can divide into manmade region and natural region. The cost image is obtained by the canny edge detector and the buffering technique is used to extract the ranging cost image. The seamline is extracted by applying the Dijkstra algorithm to a cost image generated through the labeling process of the extracted edge information. Histogram matching is also conducted to alleviate radiometric distortion between adjacent images acquired at different time. In the experimental results using the KOMPSAT-2/3 satellite imagery, it is confirmed that the proposed method greatly reduces the visual discontinuity caused by geometric difference of adjacent images and the computation time.

• Ocean Systems Engineering
• /
• v.6 no.3
• /
• pp.217-231
• /
• 2016

This paper presents a new approach for underwater image analysis using the bi-dimensional empirical mode decomposition (BEMD) technique and the phase congruency information. The BEMD algorithm, fully unsupervised, it is mainly applied to texture extraction and image filtering, which are widely recognized as a difficult and challenging machine vision problem. The phase information is the very stability feature of image. Recent developments in analysis methods on the phase congruency information have received large attention by the image researchers. In this paper, the proposed method is called the EP model that inherits the advantages of the first two algorithms, so this model is suitable for processing underwater image. Moreover, the receiver operating characteristic (ROC) curve is presented in this paper to solve the problem that the threshold is greatly affected by personal experience when underwater image edge detection is performed using the EP model. The EP images are computed using combinations of the Canny detector parameters, and the binaryzation image results are generated accordingly. The ideal EP edge feature extractive maps are estimated using correspondence threshold which is optimized by ROC analysis. The experimental results show that the proposed algorithm is able to avoid the operation error caused by manual setting of the detection threshold, and to adaptively set the image feature detection threshold. The proposed method has been proved to be accuracy and effectiveness by the underwater image processing examples.

• Journal of the Korea Society of Computer and Information
• /
• v.28 no.10
• /
• pp.67-76
• /
• 2023

Although deep learning models are making innovative achievements in the field of computer vision, the problem of vulnerability to adversarial examples continues to be raised. Adversarial examples are attack methods that inject fine noise into images to induce misclassification, which can pose a serious threat to the application of deep learning models in the real world. In this paper, we propose a model that detects adversarial examples using differences in predictive values between edge-learned classification models and underlying classification models. The simple process of extracting the edges of the objects and reflecting them in learning can increase the robustness of the classification model, and economical and efficient detection is possible by detecting adversarial examples through differences in predictions between models. In our experiments, the general model showed accuracy of {49.9%, 29.84%, 18.46%, 4.95%, 3.36%} for adversarial examples (eps={0.02, 0.05, 0.1, 0.2, 0.3}), whereas the Canny edge model showed accuracy of {82.58%, 65.96%, 46.71%, 24.94%, 13.41%} and other edge models showed a similar level of accuracy also, indicating that the edge model was more robust against adversarial examples. In addition, adversarial example detection using differences in predictions between models revealed detection rates of {85.47%, 84.64%, 91.44%, 95.47%, and 87.61%} for each epsilon-specific adversarial example. It is expected that this study will contribute to improving the reliability of deep learning models in related research and application industries such as medical, autonomous driving, security, and national defense.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2016.07a
• /
• pp.101-102
• /
• 2016

본 논문에서는 차량의 특징점들 사이의 간격과 크기의 비례식으로 자동차의 차종을 식별하는 방법을 제시한다. 자동차 관련 영상은 그 편의성을 위하여 기본 RGB모델에서 Gray색상 모델로 변환시켜 사용한다. 자동차의 배경 제거는 Canny Edge Direction을 통하여 수행하고 외곽선 검을을 통하여 원하는 특징 점을 얻는다.

• Proceedings of the Korea Multimedia Society Conference
• /
• 2002.05c
• /
• pp.282-285
• /
• 2002

의료영상에서 윤곽선의 추출은 관심영역 대한 객관적인 수치 즉 면적, 부피, 장단축의 길이 등을 분석하고 3차원 재구성을 위해 선행되어야 하는 중요한 과정이다. 현재 윤곽선 추출에 대한 않은 방법들이 개발 중에 있으나 이 방법들은 한계를 지니고 있어 더 높은 수준의 처리가 요구된다. 본 논문에서는 active contour model(snake)을 이용하여 MR뇌 영상에서 종양을 추출하였다. Snake의 에너지 최적화 문제를 dynamic programming을 사용하여 개선하였으며 canny edge detection을 이용하여 잡음에 덜 민감하도록 하였다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2019.05a
• /
• pp.629-630
• /
• 2019

Canny Edge Detection Algorithm 을 이용하여 3D 스캐너로 생성된 입체 데이터에서 2D 도면 데이터로 추출 및 변환 하는 방법을 제안한다.

• Journal of the Korea Computer Graphics Society
• /
• v.8 no.2
• /
• pp.15-22
• /
• 2002

Image segmentation methods partition an image into meaningful regions. For image composition and analysis, it is desirable for the partitioned regions to represent meaningful objects in terms of human perception and manipulation. Despite the recent progress in image understanding, however, most of the segmentation methods mainly employ low-level image features and it is still highly challenging to automatically segment an image based on high-level meaning suitable for human interpretation. The concept of HCI (Human Computer Interaction) can be applied to operator-assisted image segmentation in a manner that a human operator provides guidance to automatic image processing by interactively supplying critical information about object boundaries. Intelligent Scissors and Snakes have demonstrated the effectiveness of human-assisted segmentation [2] [1]. This paper presents a method for interactive image segmentation for more efficient and effective detection and tracking of object boundaries. The presented method is partly based on the concept of Intelligent Scissors, but employs the well-established Canny edge detector for stable edge detection. It also uses "sewing method" for including weak edges in object boundaries, and 5-direction search to promote more efficient and stable linking of neighboring edges than the previous methods.

• Journal of the Institute of Convergence Signal Processing
• /
• v.17 no.1
• /
• pp.18-25
• /
• 2016

As more people have cars, the threat of traffic accidents is posed on men and women of all ages. The main culprit of traffic accidents is driving while intoxicated or drowsy. The method to recognize and prevent the cause of traffic accidents is to use lane detection. In this study, a total of 4,000 frames (day image: 2,900 frames, night image: 1,100 frames) were used to test lane detection. According to the test, in the case of day image, when the threshold of Sobel edge detection technique was detected with second-order differential equation, there was the highest candidate lane detection rate which was 86.1%. In the threshold of Canny edge detection technique, the highest detection rate of 88.0% was found at Low=50, and High=300. In the case of night image, the threshold of Sobel edge detection technique, when horizontal calculation and vertical calculation had second-order differential equation, and when horizontal-vertical calculation had 1.5th-order differential equation, there was the highest detection rate which was 83.1%. In the threshold of Canny edge detection technique, the highest detection rate of 89.9% was found at Low=50, and High=300.

• Journal of Advanced Information Technology and Convergence
• /
• v.9 no.1
• /
• pp.115-125
• /
• 2019

In this paper, the image is received from the camera and the lane is sensed. There are various ways to detect lanes. Generally, the method of detecting edges uses a lot of the Sobel edge detection and the Canny edge detection. The minimum use of multiplication and division is used when designing for the hardware configuration. The images are tested using a black box image mounted on the vehicle. Because the top of the image of the used the black box is mostly background, the calculation process is excluded. Also, to speed up, YCbCr is calculated from the image and only the data for the desired color, white and yellow lane, is obtained to detect the lane. The median filter is used to remove noise from images. Intermediate filters excel at noise rejection, but they generally take a long time to compare all values. In this paper, by using addition, the time can be shortened by obtaining and using the result value of the median filter. In case of the Sobel edge detection, the speed is faster and noise sensitive compared to the Canny edge detection. These shortcomings are constructed using complementary algorithms. It also organizes and processes data into parallel processing pipelines. To reduce the size of memory, the system does not use memory to store all data at each step, but stores it using four line buffers. Three line buffers perform mask operations, and one line buffer stores new data at the same time as the operation. Through this work, memory can use six times faster the processing speed and about 33% greater quantity than other methods presented in this paper. The target operating frequency is designed so that the system operates at 50MHz. It is possible to use 2157fps for the images of 640by360 size based on the target operating frequency, 540fps for the HD images and 240fps for the Full HD images, which can be used for most images with 30fps as well as 60fps for the images with 60fps. The maximum operating frequency can be used for larger amounts of the frame processing.