• 전기학회논문지
• /
• 제65권1호
• /
• pp.142-149
• /
• 2016

This paper proposes a method to detect LP(License Plate) of vehicles in indoor and outdoor parking lots. In restricted environment, there are many conventional methods for detecting LP. But, it is difficult to detect LP in natural and complex scenes with background clutters because several patterns similar with text or LP always exist in complicated backgrounds. To verify the performance of LP text detection in natural images, we apply MB-LGP feature by combining with ensemble machine learning algorithm in purpose of selecting optimal features of small number in huge pool. The feature selection is performed by adaptive boosting algorithm that shows great performance in minimum false positive detection ratio and in computing time when combined with cascade approach. MSER is used to provide initial text regions of vehicle LP. Throughout the experiment using real images, the proposed method functions robustly extracting LP in natural scene as well as the controlled environment.

• International Journal of Contents
• /
• 제6권1호
• /
• pp.6-11
• /
• 2010

Recently, there has been research to use portable digital camera to recognize objects in natural scene images, including labels or marks on a cylindrical surface. In many cases, text or logo in a label can be distorted by a structural movement of the object on which the label resides. Since the distortion in the label can degrade the performance of object recognition, the label should be rectified or restored from deformations. In this paper, a new method for label detection and restoration in digital images is presented. In the detection phase, the Hough transform is employed to detect two vertical boundaries of the label, and a horizontal edge profile is analyzed to detect upper-side and lower-side boundaries of the label. Then, the biquadratic transformation is used to restore the rectangular shape of the label. The proposed algorithm performs restoration of 3D objects in a 2D space, and it requires neither an auxiliary hardware such as 3D camera to construct 3D models nor a multi-camera to capture objects in different views. Experimental results demonstrate the effectiveness of the proposed method.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제14권10호
• /
• pp.4117-4135
• /
• 2020

Due to the angle of view, illumination and scene diversity, real-time dynamic face detection and recognition is no small difficulty in those unrestricted environments. In this study, we used the intrinsic correlation between detection and calibration, using a multi-task cascaded convolutional neural network(MTCNN) to improve the efficiency of face recognition, and the output of each core network is mapped in parallel to a compact Euclidean space, where distance represents the similarity of facial features, so that the target face can be identified as quickly as possible, without waiting for all network iteration calculations to complete the recognition results. And after the angle of the target face and the illumination change, the correlation between the recognition results can be well obtained. In the actual application scenario, we use a multi-camera real-time monitoring system to perform face matching and recognition using successive frames acquired from different angles. The effectiveness of the method was verified by several real-time monitoring experiments, and good results were obtained.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2000년도 하계학술대회 논문집 D
• /
• pp.2814-2818
• /
• 2000

This paper describes an image processing algorithm capable of recognizing the road lane using a CDF (Cumulative Distribution Function). which is designed for the model function of the road lane. The CDF has distinctive peak points at the vicinity of the lane direction because of the directional and positional continuities of the lane. We construct a scatter diagram by collecting the edge pixels with the direction corresponding to the peak point of the CDF and carry out the principal axis-based line fitting for the scatter diagram to obtain the lane information. As noises play the role of making a lot of similar features to the lane appear and disappear in the image we introduce a recursive estimator of the function to reduce the noise effect and a scene understanding index (SUI) formulated by statistical parameters of the CDF to prevent a false alarm or miss detection. The proposed algorithm has been implemented in a real time on the video data obtained from a test vehicle driven in a typical highway.

• Journal of KIEE
• /
• 제11권1호
• /
• pp.69-77
• /
• 2001

This paper describes an image processing algorithm capable of recognizing road lanes by using a CDF(cumulative distribution function). The CDF is designed for the model function of road lanes. Based on the assumptions that there are no abrupt changes in the direction and location of road lanes and that the intensity of lane boundaries differs from that of the background, we formulated the CDF, which accumulates the edge magnitude for edge directions. The CDF has distinctive peak points at the vicinity of lane directions due to the directional and the positional continuities of a lane. To obtain lane-related information a scatter diagram was constructed by collecting edge pixels, of which the direction corresponds to the peak point of the CDF, then the principal axis-based line fitting was performed for the scatter diagram. Noises can cause many similar features to appear and to disappear in an image. Therefore, to reduce the noise effect a recursive estimator of the CDF was introduced, and also to prevent false alarms or miss detection a scene understanding index (DUI) was formulated by the statistical parameters of the CDF. The proposed algorithm has been implemented in real time on video data obtained from a test vehicle driven on a typical highway.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제11권2호
• /
• pp.118-123
• /
• 2011

Video based object tracking normally deals with non-stationary image streams that change over time. Robust and real time moving object tracking is considered to be a problematic issue in computer vision. Multiple object tracking has many practical applications in scene analysis for automated surveillance. In this paper, we introduce a specified object tracking based particle filter used in an environment of multiple moving objects. A differential image region based tracking method for the detection of multiple moving objects is used. In order to ensure accurate object detection in an unconstrained environment, a background image update method is used. In addition, there exist problems in tracking a particular object through a video sequence, which cannot rely only on image processing techniques. For this, a probabilistic framework is used. Our proposed particle filter has been proved to be robust in dealing with nonlinear and non-Gaussian problems. The particle filter provides a robust object tracking framework under ambiguity conditions and greatly improves the estimation accuracy for complicated tracking problems.

• 로봇학회논문지
• /
• 제18권3호
• /
• pp.346-351
• /
• 2023

With the release of numerous open driving datasets, the demand for domain adaptation in perception tasks has increased, particularly when transferring knowledge from rich datasets to novel domains. However, it is difficult to solve the change 1) in the sensor domain caused by heterogeneous LiDAR sensors and 2) in the environmental domain caused by different environmental factors. We overcome domain differences in the semi-supervised setting with 3-stage model parameter training. First, we pre-train the model with the source dataset with object scaling based on statistics of the object size. Then we fine-tine the partially frozen model weights with copy-and-paste augmentation. The 3D points in the box labels are copied from one scene and pasted to the other scenes. Finally, we use the knowledge distillation method to update the student network with a moving average from the teacher network along with a self-training method with pseudo labels. Test-Time Augmentation with varying z values is employed to predict the final results. Our method achieved 3rd place in ECCV 2022 workshop on the 3D Perception for Autonomous Driving challenge.

• 국제학술발표논문집
• /
• The 1th International Conference on Construction Engineering and Project Management
• /
• pp.539-543
• /
• 2005

The research presented in this paper enables real-time 3D modeling to help make construction processes ultimately faster, more predictable and safer. Initial research efforts used an emerging sensor technology and proved its usefulness in the acquisition of range information for the detection and efficient representation of static and moving objects. Based on the time-of-flight principle, the sensor acquires range and intensity information of each image pixel within the entire sensor's field-of-view in real-time with frequencies of up to 30 Hz. However, real-time working range data processing algorithms need to be developed to rapidly process range information into meaningful 3D computer models. This research ultimately focuses on the application of safer heavy equipment operation. The paper compares (a) a previous research effort in convex hull modeling using sparse range point clouds from a single laser beam range finder, to (b) high-frame rate update Flash LADAR (Laser Detection and Ranging) scanning for complete scene modeling. The presented research will demonstrate if the FlashLADAR technology can play an important role in real-time modeling of infrastructure assets in the near future.

• 스마트미디어저널
• /
• 제11권7호
• /
• pp.94-103
• /
• 2022

Weeds bring disadvantages to crops since they can damage them, and a clean treatment with less pollution and contamination should be developed. Artificial intelligence gives new hope to agriculture to achieve smart farming. This study delivers an automated weeds growth point detection using deep learning. This study proposes a combination of semantic graphics for generating data annotation and U-Net with pre-trained deep learning as a backbone for locating the growth point of the weeds on the given field scene. The dataset was collected from an actual field. We measured the intersection over union, f1-score, precision, and recall to evaluate our method. Moreover, Mobilenet V2 was chosen as the backbone and compared with Resnet 34. The results showed that the proposed method was accurate enough to detect the growth point and handle the brightness variation. The best performance was achieved by Mobilenet V2 as a backbone with IoU 96.81%, precision 97.77%, recall 98.97%, and f1-score 97.30%.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2023년도 추계학술발표대회
• /
• pp.637-640
• /
• 2023

This paper presents multimodal image fusion with human pose for detecting abnormal human behaviors in low illumination conditions. Detecting human behaviors in low illumination conditions is challenging due to its limited visibility of the objects of interest in the scene. Multimodal image fusion simultaneously combines visual information in the visible spectrum and thermal radiation information in the long-wave infrared spectrum. We propose an abnormal event detection scheme based on the multimodal fused image and the human poses using the keypoints to characterize the action of the human body. Our method assumes that human behaviors are well correlated to body keypoints such as shoulders, elbows, wrists, hips. In detail, we extracted the human keypoint coordinates from human targets in multimodal fused videos. The coordinate values are used as inputs to train a multilayer perceptron network to classify human behaviors as normal or abnormal. Our experiment demonstrates a significant result on multimodal imaging dataset. The proposed model can capture the complex distribution pattern for both normal and abnormal behaviors.