• 한국측량학회지
• /
• 제38권6호
• /
• pp.635-644
• /
• 2020

딥러닝(DL)을 이용한 객체인식, 탐지 및 분할하는 연구는 여러 분야에서 활용되고 있으며, 주로 영상을 DL 모델의 학습 데이터로 사용하고 있지만, 본 논문은 영상뿐 아니라 공간정보 특성을 포함하는 다양한 학습 데이터(multimodal training data)를 향상된 영역기반 합성곱 신경망(R-CNN)인 Detectron2 모델 학습에 사용하여 객체를 분할하고 건물을 탐지하는 것이 목적이다. 이를 위하여 적외선 항공영상과 라이다 데이터의 내재된 객체의 윤곽 및 통계적 질감정보인 Haralick feature와 같은 여러 특성을 추출하였다. DL 모델의 학습 성능은 데이터의 수량과 특성뿐 아니라 융합방법에 의해 좌우된다. 초기융합(early fusion)과 후기융합(late fusion)의 혼용방식인 하이브리드 융합(hybrid fusion)을 적용한 결과 33%의 건물을 추가적으로 탐지 할 수 있다. 이와 같은 실험 결과는 서로 다른 특성 데이터의 복합적 학습과 융합에 의한 상호보완적 효과를 입증하였다고 판단된다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제15권12호
• /
• pp.4439-4455
• /
• 2021

Compared with two-stage object detection algorithms, one-stage algorithms provide a better trade-off between real-time performance and accuracy. However, these methods treat the intermediate features equally, which lacks the flexibility to emphasize meaningful information for classification and location. Besides, they ignore the interaction of contextual information from different scales, which is important for medium and small objects detection. To tackle these problems, we propose an image pyramid network based on dual attention mechanism (DAIPNet), which builds an image pyramid to enrich the spatial information while emphasizing multi-scale informative features based on dual attention mechanisms for one-stage object detection. Our framework utilizes a pre-trained backbone as standard detection network, where the designed image pyramid network (IPN) is used as auxiliary network to provide complementary information. Here, the dual attention mechanism is composed of the adaptive feature fusion module (AFFM) and the progressive attention fusion module (PAFM). AFFM is designed to automatically pay attention to the feature maps with different importance from the backbone and auxiliary network, while PAFM is utilized to adaptively learn the channel attentive information in the context transfer process. Furthermore, in the IPN, we build an image pyramid to extract scale-wise features from downsampled images of different scales, where the features are further fused at different states to enrich scale-wise information and learn more comprehensive feature representations. Experimental results are shown on MS COCO dataset. Our proposed detector with a 300 × 300 input achieves superior performance of 32.6% mAP on the MS COCO test-dev compared with state-of-the-art methods.

• 한국지능시스템학회:학술대회논문집
• /
• 한국지능시스템학회 2007년도 추계학술대회 학술발표 논문집
• /
• pp.11-14
• /
• 2007

인간과 컴퓨터간의 상호교류 하는데 있어서 감정 인식은 필수라 하겠다. 그래서 본 논문에서는 음성 신호 및 얼굴 영상을 BL(Bayesian Learning)과 PCA(Principal Component Analysis)에 적용하여 5가지 감정 (Normal, Happy, Sad, Anger, Surprise) 으로 패턴 분류하였다. 그리고 각각 신호의 단점을 보완하고 인식률을 높이기 위해 결정 융합 방법과 특징 융합 방법을 이용하여 감정융합을 실행하였다. 결정 융합 방법은 각각 인식 시스템을 통해 얻어진 인식 결과 값을 퍼지 소속 함수에 적용하여 감정 융합하였으며, 특정 융합 방법은 SFS(Sequential Forward Selection)특정 선택 방법을 통해 우수한 특정들을 선택한 후 MLP(Multi Layer Perceptron) 기반 신경망(Neural Networks)에 적용하여 감정 융합을 실행하였다.

• 한국멀티미디어학회논문지
• /
• 제13권6호
• /
• pp.875-883
• /
• 2010

In this paper, we present a secure biometric hashing scheme for face recognition by random fusion of global and local features. The Fourier-Mellin transform and Radon transform are adopted respectively to form specialized representation of global and local features, due to their invariance to geometric operations. The final biometric hash is securely generated by random weighting sum of both feature sets. A fourfold key is involved in our algorithm to ensure the security and privacy of biometric templates. The proposed biometric hash can be revocable and replaced by using a new key. Moreover, the attacker cannot obtain any information about the original biometric template without knowing the secret key. The experimental results confirm that our scheme has a satisfactory accuracy performance in terms of EER.

• 한국측량학회지
• /
• 제32권6호
• /
• pp.581-590
• /
• 2014

With rapid land development, land category should be updated on a regular basis. However, manual field surveys have certain limitations. In this study, attempts were made to extract a feature vector considering spectral signature by parcel, PIMP (Percent Imperviousness), texture, and VIs (Vegetation Indices) based on RapidEye satellite image and cadastral map. A total of nine land categories in which feature vectors were significantly extracted from the images were selected and classified using SVM (Support Vector Machine). According to accuracy assessment, by comparing the cadastral map and classification result, the overall accuracy was 0.74. In the paddy-field category, in particular, PO acc. (producer's accuracy) and US acc. (user's accuracy) were highest at 0.85 and 0.86, respectively.

• ETRI Journal
• /
• 제41권6호
• /
• pp.750-759
• /
• 2019

Presently, the extraction of hand-crafted features is still the dominant method in radar emitter recognition. To solve the complicated problems of selection and updation of empirical features, we present a novel automatic feature extraction structure based on deep learning. In particular, a convolutional neural network (CNN) is adopted to extract high-level abstract representations from the time-frequency images of emitter signals. Thus, the redundant process of designing discriminative features can be avoided. Furthermore, to address the performance degradation of a single platform, we propose the construction of an ensemble learning-based architecture for multi-platform fusion recognition. Experimental results indicate that the proposed algorithms are feasible and effective, and they outperform other typical feature extraction and fusion recognition methods in terms of accuracy. Moreover, the proposed structure could be extended to other prevalent ensemble learning alternatives.

• Journal of Information Processing Systems
• /
• 제14권1호
• /
• pp.218-227
• /
• 2018

Vocal effort detection is important for both robust speech recognition and speaker recognition. In this paper, the spectral information entropy feature which contains more salient information regarding the vocal effort level is firstly proposed. Then, the model fusion method based on complementary model is presented to recognize vocal effort level. Experiments are conducted on isolated words test set, and the results show the spectral information entropy has the best performance among the three kinds of features. Meanwhile, the recognition accuracy of all vocal effort levels reaches 81.6%. Thus, potential of the proposed method is demonstrated.

• 한국멀티미디어학회논문지
• /
• 제24권1호
• /
• pp.1-12
• /
• 2021

In this paper, we present a new architecture for semantic segmentation. Semantic segmentation aims at a pixel-wise classification which is important to fully understand images. Previous semantic segmentation networks use features of multi-layers in the encoder to predict final results. However, they do not contain various receptive fields in the multi-layers features, which easily lead to inaccurate results for boundaries between different classes and small objects. To solve this problem, we propose a multi-path feature fusion module that allows for features of each layers to contain various receptive fields by use of a set of dilated convolutions with different dilatation rates. Various experiments demonstrate that our method outperforms previous methods in terms of mean intersection over unit (mIoU).

• 한국인공지능학회지
• /
• 제11권3호
• /
• pp.17-22
• /
• 2023

In this paper, we propose a black ice detection system using Convolutional Neural Networks (CNNs). Black ice poses a serious threat to road safety, particularly during winter conditions. To overcome this problem, we introduce a CNN-based architecture for real-time black ice detection with an encoder-decoder network, specifically designed for real-time black ice detection using thermal images. To train the network, we establish a specialized experimental platform to capture thermal images of various black ice formations on diverse road surfaces, including cement and asphalt. This enables us to curate a comprehensive dataset of thermal road black ice images for a training and evaluation purpose. Additionally, in order to enhance the accuracy of black ice detection, we propose a multi-scale dilation convolution feature fusion (MsDC-FF) technique. This proposed technique dynamically adjusts the dilation ratios based on the input image's resolution, improving the network's ability to capture fine-grained details. Experimental results demonstrate the superior performance of our proposed network model compared to conventional image segmentation models. Our model achieved an mIoU of 95.93%, while LinkNet achieved an mIoU of 95.39%. Therefore, it is concluded that the proposed model in this paper could offer a promising solution for real-time black ice detection, thereby enhancing road safety during winter conditions.

• 한국멀티미디어학회논문지
• /
• 제22권12호
• /
• pp.1457-1465
• /
• 2019

3D convolution is to stack multiple consecutive frames to form a cube, and then apply the 3D convolution kernel in the cube. In this structure, each feature map of the convolutional layer is connected to multiple adjacent sequential frames in the previous layer, thus capturing the motion information. However, due to the changes of pedestrian posture, motion and position, the convolution at the same place is inappropriate, and when the 3D convolution kernel is convoluted in the time domain, only time domain features of three consecutive frames can be extracted, which is not a good enough to get action information. This paper proposes an action recognition method based on feature fusion of 3D convolutional neural network. Based on the VGG16 network model, sending a pre-acquired optical flow image for learning, then get the time domain features, and then the feature of the time domain is extracted from the features extracted by the 3D convolutional neural network. Finally, the behavior classification is done by the SVM classifier.