• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제15권10호
• /
• pp.3729-3749
• /
• 2021

At present, deep convolution network-based salient object detection (SOD) has achieved impressive performance. However, it is still a challenging problem to make full use of the multi-scale information of the extracted features and which appropriate feature fusion method is adopted to process feature mapping. In this paper, we propose a new adjacency auxiliary network (AANet) based on multi-scale feature fusion for SOD. Firstly, we design the parallel connection feature enhancement module (PFEM) for each layer of feature extraction, which improves the feature density by connecting different dilated convolution branches in parallel, and add channel attention flow to fully extract the context information of features. Then the adjacent layer features with close degree of abstraction but different characteristic properties are fused through the adjacent auxiliary module (AAM) to eliminate the ambiguity and noise of the features. Besides, in order to refine the features effectively to get more accurate object boundaries, we design adjacency decoder (AAM_D) based on adjacency auxiliary module (AAM), which concatenates the features of adjacent layers, extracts their spatial attention, and then combines them with the output of AAM. The outputs of AAM_D features with semantic information and spatial detail obtained from each feature are used as salient prediction maps for multi-level feature joint supervising. Experiment results on six benchmark SOD datasets demonstrate that the proposed method outperforms similar previous methods.

• 인터넷정보학회논문지
• /
• 제15권6호
• /
• pp.117-124
• /
• 2014

모바일 기기를 사용한 실시간 비디오 영상처리분야의 중요 객체탐색 및 추적의 문제에 있어서 난제는 복잡한 배경속에서 전경을 구분해 내는 일이다. 본 논문에서는 기계학습을 위한 특성벡터 선정의 문제를 위한 문맥인식 모델을 제시하여 잡음제거를 위한 기계학습기반의 구분자를 구현하였다. 수학적으로 NP-hard로 알려진 가장 가까운 이웃을 사용한 문맥인식 특성벡터 선정 알고리즘의 구현에 있어서, 본 논문은 연산횟수를 줄인 유사방법론에 대해 자세히 거론하였다. 또한, 문맥인식 성격을 가미한 특성벡터 선정을 통해 얻어진 특성 공간에서의 향상된 분리성에 대해 주성분 분석을 통해 엄밀한 분석결과를 제시하였다. 전반적인 성능 향상의 정도를 계측하기 위해 다양한 기계학습 방법론, 예를 들어, 다층신경망, 지원벡터기계, 나이브베이지안, 회귀분석 등을 사용해 비교결과를 제시하였다. 본 논문에서 제시한 방법론의 성능과 계산상 자원사용에 대한 내용을 결론으로 서술하였다.

• 한국융합학회논문지
• /
• 제11권12호
• /
• pp.39-47
• /
• 2020

본 논문에서는 이미지에서 돌출되는 객체를 검출하기 위해 Saliency Attention을 이용한 딥러닝 기반의 검출 방법을 제안하였다. 돌출 객체 검출은 사람의 시선이 집중되는 물체를 배경으로부터 분리시키는 것이며, 이미지에서 관련성이 높은 부분을 결정한다. 객체 추적 및 검출, 인식 등의 다양한 분야에서 유용하게 사용된다. 기존의 딥러닝 기반 방법들은 대부분 오토인코더 구조로, 특징을 압축 및 추출하는 인코더와 추출된 특징을 복원 및 확장하는 디코더에서 많은 특징 손실이 발생한다. 이러한 손실로 돌출 객체 영역에 손실이 발생하거나 배경을 객체로 검출하는 문제가 있다. 제안하는 방법은 오토인코더 구조에서 특징 손실을 감소시키고 배경 영역을 억제하기 위해 Saliency Attention을 제안하였다. ELU 활성화 함수를 이용해 특징 값의 영향력을 결정하며 각각 정규화된 음수 및 양수 영역의 특징값에 Attention을 진행하였다. 제안하는 Attention 기법을 통해 배경 영역을 억제하며 돌출 객체 영역을 강조하였다. 실험 결과에서는 제안하는 방법이 기존 방법과 비교하여 향상된 검출 결과를 보였다.

• International journal of advanced smart convergence
• /
• 제9권3호
• /
• pp.85-96
• /
• 2020

It is important to extract salient object image and to solve the invariance problem for image recognition. In this paper we propose a new hybrid algorithm for invariance and improved classification performance in image recognition, whose algorithm is combined by FT(Frequency-tuned Salient Region Detection) algorithm, Guided filter, Zernike moments, and a simple artificial neural network (Multi-layer Perceptron). The conventional FT algorithm is used to extract initial salient object image, the guided filtering to preserve edge details, Zernike moments to solve invariance problem, and a classification to recognize the extracted image. For guided filtering, guided filter is used, and Multi-layer Perceptron which is a simple artificial neural networks is introduced for classification. Experimental results show that this algorithm can achieve a superior performance in the process of extracting salient object image and invariant moment feature. And the results show that the algorithm can also classifies the extracted object image with improved recognition rate.

• Computers and Concrete
• /
• 제26권5호
• /
• pp.411-420
• /
• 2020

Deep Convolution neural network (DCNN) has been widely used in the healthy maintenance of civil infrastructure. Using DCNN to improve crack detection performance has attracted many researchers' attention. In this paper, a light-weight spatial attention network module is proposed to strengthen the representation capability of ResNet and improve the crack detection performance. It utilizes attention mechanism to strengthen the interested objects in global receptive field of ResNet convolution layers. Global average spatial information over all channels are used to construct an attention scalar. The scalar is combined with adaptive weighted sigmoid function to activate the output of each channel's feature maps. Salient objects in feature maps are refined by the attention scalar. The proposed spatial attention module is stacked in ResNet50 to detect crack. Experiments results show that the proposed module can got significant performance improvement in crack detection.

• 대한원격탐사학회:학술대회논문집
• /
• 대한원격탐사학회 2003년도 Proceedings of ACRS 2003 ISRS
• /
• pp.408-410
• /
• 2003

In this paper, an improved gradient direction assisted linking algorithm is proposed. This algorithm begins with initial seeds satisfying some local criteria. Then it will search along the direction provided by the initial point. A window will be generated in the gradient direction of the current point. Instead of the conventional method which only considers the value of the local salient structure, an improved mathematical model is proposed to describe the desired linear features. This model not only considers the value of the salient structure but also the direction of it. Furthermore, the linking problem under this model can be efficiently solved by dynamic programming method. This algorithm is tested for linear features detection in IKONOS images. The result demonstrates this algorithm is quite promising.

• 전자공학회논문지
• /
• 제51권11호
• /
• pp.181-186
• /
• 2014

본 논문에서는 보행자 상반신 검출기의 성능을 향상하기 위한 세그먼테이션에 기반한 특징 추출 방법을 제안한다. 상반신의 부분별 색상 분포를 활용한 멀티 파트 컬러 세그먼테이션을 사용하여 국소 특징이 갖는 한계로 인해 발생하는 오검출의 감소에 효과적인 "전역적" 윤곽 특징을 추출한다. 컬러 공간과 히스토그램 분해도에 따른 성능을 분석하였으며, 자체 구축한 보행자 상반신 영상을 사용한 실험을 통해서 제안한 방법으로 추출한 특징이 국소 특징 기반 검출기의 오검출 감소에 효과적임을 확인하였다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제7권8호
• /
• pp.1843-1859
• /
• 2013

The continuous increase of the spatial resolution of remote sensing images brings great challenge to image analysis and processing. Traditional prior knowledge-based region detection and target recognition algorithms for processing high resolution remote sensing images generally employ a global searching solution, which results in prohibitive computational complexity. In this paper, a more efficient region of interest (ROI) detection algorithm based on visual attention and threshold segmentation (VA-TS) is proposed, wherein a visual attention mechanism is used to eliminate image segmentation and feature detection to the entire image. The input image is subsampled to decrease the amount of data and the discrete moment transform (DMT) feature is extracted to provide a finer description of the edges. The feature maps are combined with weights according to the amount of the "strong points" and the "salient points". A threshold segmentation strategy is employed to obtain more accurate region of interest shape information with the very low computational complexity. Experimental statistics have shown that the proposed algorithm is computational efficient and provide more visually accurate detection results. The calculation time is only about 0.7% of the traditional Itti's model.

• 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.

• ETRI Journal
• /
• 제37권6호
• /
• pp.1220-1230
• /
• 2015

An autonomous valet parking (AVP) system is designed to locate a vacant parking space and park the vehicle in which it resides on behalf of the driver, once the driver has left the vehicle. In addition, the AVP is able to direct the vehicle to a location desired by the driver when requested. In this paper, for an AVP system, we introduce technology to recognize a parking space using image sensors. The proposed technology is mainly divided into three parts. First, spatial analysis is carried out using a height map that is based on dense motion stereo. Second, modelling of road markings is conducted using a probability map with a new salient-line feature extractor. Finally, parking space recognition is based on a Bayesian classifier. The experimental results show an execution time of up to 10 ms and a recognition rate of over 99%. Also, the performance and properties of the proposed technology were evaluated with a variety of data. Our algorithms, which are part of the proposed technology, are expected to apply to various research areas regarding autonomous vehicles, such as map generation, road marking recognition, localization, and environment recognition.