• Journal of Institute of Control, Robotics and Systems
• /
• v.7 no.8
• /
• pp.656-663
• /
• 2001

This paper presents a sensor fusion method to recognize a cylindrical object a CCD camera, a laser slit beam and ultrasonic sensors on a pan/tilt device. For object recognition with a vision sensor, an active light source projects a stripe pattern of light on the object surface. The 2D image data are transformed into 3D data using the geometry between the camera and the laser slit beam. The ultrasonic sensor uses an ultrasonic transducer array mounted in horizontal direction on the pan/tilt device. The time of flight is estimated by finding the maximum correlation between the received ultrasonic pulse and a set of stored templates - also called a matched filter. The distance of flight is calculated by simply multiplying the time of flight by the speed of sound and the maximum amplitude of the filtered signal is used to determine the face angle to the object. To determine the position and the radius of cylindrical objects, we use a statistical sensor fusion. Experimental results show that the fused data increase the reliability for the object recognition.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.18 no.8
• /
• pp.2103-2118
• /
• 2024

Chinese herbal slices (CHS) automated recognition based on computer vision plays a critical role in the practical application of intelligent Chinese medicine. Due to the complexity and similarity of herbal images, identifying Chinese herbal slices is still a challenging task. Especially, easily-confused CHS have higher inter-class and intra-class complexity and similarity issues, the existing deep learning models are less adaptable to identify them efficiently. To comprehensively address these problems, a novel tiny easily-confused CHS dataset has been built firstly, which includes six pairs of twelve categories with about 2395 samples. Furthermore, we propose a ResNeSt-CHS model that combines multilevel perception fusion (MPF) and perceptive sparse fusion (PSF) blocks for efficiently recognizing easilyconfused CHS images. To verify the superiority of the ResNeSt-CHS and the effectiveness of our dataset, experiments have been employed, validating that the ResNeSt-CHS is optimal for easily-confused CHS recognition, with 2.1% improvement of the original ResNeSt model. Additionally, the results indicate that ResNeSt-CHS is applied on a relatively small-scale dataset yet high accuracy. This model has obtained state-of-the-art easily-confused CHS classification performance, with accuracy of 90.8%, far beyond other models (EfficientNet, Transformer, and ResNeSt, etc) in terms of evaluation criteria.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.13 no.3
• /
• pp.461-469
• /
• 2010

In this paper, we have proposed an improved fusion method of detection features which can enhance the detection probability under the given false alarm rate in the prescreening stage of SAR ATR(Synthetic Aperture Radar Automatic Target Recognition) system. Since the detection features have the positive correlation, the detection performance can be improved if the joint probability distribution of detection features is considered in the fusion process. The detection region is designed as a simple piecewise linear function which can be represented by few parameters. The parameters for the detection region can be derived by training the sample SAR images to maximize the detection probability with the given false alarm rate. Simulation result shows that the detection performance of the proposed method is improved for all combinations of detection features.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.05a
• /
• pp.493-495
• /
• 2023

Deep learning methods and attention mechanisms have been incorporated to improve facial emotion recognition, which has recently attracted much attention. The fusion approaches have improved accuracy by combining various types of information. This research proposes a fusion network with self-attention and local attention mechanisms. It uses a multi-layer perceptron network. The network extracts distinguishing characteristics from facial images using pre-trained models on RAF-DB dataset. We outperform the other fusion methods on RAD-DB dataset with impressive results.

• Journal of Broadcast Engineering
• /
• v.19 no.2
• /
• pp.166-173
• /
• 2014

Recently, human action recognition have been developed for various broadcasting and video process. Since a video can consist of various scenes, keypoint approaches have been more attracted than template based methods for real application. Keypoint approahces tried to find regions having motion in video, and made 3-dimensional patches. Then, descriptors using histograms were computed from the patches, and a classifier based on machine learning method was applied to detect actions in video. However, a single classifier was difficult to handle various human actions. In order to improve this problem, approaches using multi classifiers were used to detect and to recognize objects. Thus, we propose a new human action recognition using decision-level fusion with support vector machine and sparse representation. The proposed method extracted descriptors based on keypoint approach from a video, and acquired results from each classifier for human action recognition. Then, we applied weights which were acquired by training stage to fuse each results from two classifiers. The experiment results in this paper show better result than a previous fusion method.

• Proceedings of the KIEE Conference
• /
• 2006.04a
• /
• pp.123-125
• /
• 2006

This paper discusses the method of vision based sensor fusion system for biped robot walking. Most researches on biped walking robot have mostly focused on walking algorithm itself. However, developing vision systems for biped walking robot is an important and urgent issue since biped walking robots are ultimately developed not only for researches but to be utilized in real life. In the research, systems for environment recognition and tole-operation have been developed for task assignment and execution of biped robot as well as for human robot interaction (HRI) system. For carrying out certain tasks, an object tracking system using modified optical flow algorithm and obstacle recognition system using enhanced template matching and hierarchical support vector machine algorithm by wireless vision camera are implemented with sensor fusion system using other sensors installed in a biped walking robot. Also systems for robot manipulating and communication with user have been developed for robot.

• Proceedings of the KIEE Conference
• /
• 1992.07a
• /
• pp.417-421
• /
• 1992

Ultrasonic and vision data need to be fused for efficient object recognition, especially in mobile robot navigation. In the proposed approach, the whole ultrasonic echo signal is utilized and data fusion is performed based on each sensor's characteristic. It is shown to be effective through the experiment results.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.25 no.5
• /
• pp.437-443
• /
• 2015

This paper presents face recognition based on the fusion of visible image and thermal infrared (IR) texture estimated from the face image in the visible spectrum. The proposed face recognition scheme uses a multi- layer neural network to estimate thermal texture from visible imagery. In the training process, a set of visible and thermal IR image pairs are used to determine the parameters of the neural network to learn a complex mapping from a visible image to its thermal texture in the low-dimensional feature space. The trained neural network estimates the principal components of the thermal texture corresponding to the input visible image. Extensive experiments on face recognition were performed using two popular face recognition algorithms, Eigenfaces and Fisherfaces for NIST/Equinox database for benchmarking. The fusion of visible image and thermal IR texture demonstrated improved face recognition accuracies over conventional face recognition in terms of receiver operating characteristics (ROC) as well as first matching performances.

• Journal of KIISE:Software and Applications
• /
• v.31 no.7
• /
• pp.929-938
• /
• 2004

This paper proposes a pose-invariant face recognition method using cylindrical model and stereo camera. We divided this paper into two parts. One is single input image case, the other is stereo input image case. In single input image case, we normalized a face's yaw pose using cylindrical model, and in stereo input image case, we normalized a face's pitch pose using cylindrical model with previously estimated pitch pose angle by the stereo geometry. Also, since we have an advantage that we can utilize two images acquired at the same time, we can increase overall recognition performance by decision-level fusion. Through representative experiments, we achieved an increased recognition rate from 61.43% to 94.76% by the yaw pose transform, and the recognition rate with the proposed method achieves as good as that of the more complicated 3D face model. Also, by using stereo camera system we achieved an increased recognition rate 5.24% more for the case of upper face pose, and 3.34% more by decision-level fusion.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.33 no.6
• /
• pp.1099-1110
• /
• 2023

The recognition system of autonomous driving and robot navigation performs vision work such as object recognition, tracking, and lane detection after multi-sensor fusion to improve performance. Currently, research on a deep learning model based on the fusion of a camera and a lidar sensor is being actively conducted. However, deep learning models are vulnerable to adversarial attacks through modulation of input data. Attacks on the existing multi-sensor-based autonomous driving recognition system are focused on inducing obstacle detection by lowering the confidence score of the object recognition model.However, there is a limitation that an attack is possible only in the target model. In the case of attacks on the sensor fusion stage, errors in vision work after fusion can be cascaded, and this risk needs to be considered. In addition, an attack on LIDAR's point cloud data, which is difficult to judge visually, makes it difficult to determine whether it is an attack. In this study, image scaling-based camera-lidar We propose an attack method that reduces the accuracy of LCCNet, a fusion model (camera-LiDAR calibration model). The proposed method is to perform a scaling attack on the point of the input lidar. As a result of conducting an attack performance experiment by size with a scaling algorithm, an average of more than 77% of fusion errors were caused.