A new algorithm for bottom-up saliency estimation is proposed.Based on the sparse coding model,a power spectral filter is proposed to eliminate the second-order residual correlation,which suppresses the global repeate...A new algorithm for bottom-up saliency estimation is proposed.Based on the sparse coding model,a power spectral filter is proposed to eliminate the second-order residual correlation,which suppresses the global repeated items effectively.In addition,aiming at modeling the mechanism of the human retina prior response to high-contrast stimuli,the effect of color context is considered.Experiments on the three publicly available databases and some psychophysical images show that the proposed model is comparable with the state-of-the-art saliency models,which not only highlights the salient objects in a complex environment but also pops up them uniformly.展开更多
The goal of salient object detection is to estimate the regions which are most likely to attract human's visual attention. As an important image preprocessing procedure to reduce the computational complexity, sali...The goal of salient object detection is to estimate the regions which are most likely to attract human's visual attention. As an important image preprocessing procedure to reduce the computational complexity, salient object detection is still a challenging problem in computer vision. In this paper, we proposed a salient object detection model by integrating local and global superpixel contrast at multiple scales. Three features are computed to estimate the saliency of superpixel. Two optimization measures are utilized to refine the resulting saliency map. Extensive experiments with the state-of-the-art saliency models on four public datasets demonstrate the effectiveness of the proposed model.展开更多
针对红外图像与可见光图像融合中细节丢失严重,红外图像的特征信息未能突出显示以及源图像的语义信息被忽视的问题,提出一种基于二次图像分解的红外图像与可见光图像融合网络(Secondary Image Decomposition For Infrared And Visible I...针对红外图像与可见光图像融合中细节丢失严重,红外图像的特征信息未能突出显示以及源图像的语义信息被忽视的问题,提出一种基于二次图像分解的红外图像与可见光图像融合网络(Secondary Image Decomposition For Infrared And Visible Image Fusion,SIDFuse)。利用编码器对源图像进行二次分解以提取不同尺度的特征信息,然后利用双元素注意力为不同尺度的特征信息分配权重、引入全局语义支路,再采用像素相加法作为融合策略,最后通过解码器重建融合图像。实验选择FLIR数据集用于训练,采用TNO和RoadScene两个数据集进行测试,并选取八种图像融合客观评价参数进行实验对比分析。由TNO数据集的图像融合实验表明,在信息熵、标准差、空间频率、视觉保真度、平均梯度、差异相关系数、多层级结构相似性、梯度融合性能评价指标上,SIDFuse比基于卷积网络中经典融合算法DenseFuse分别平均提高12.2%,9.0%,90.2%,13.9%,85.1%,16.8%,6.7%,30.7%,比最新的融合网络LRRNet分别平均提高2.5%,5.6%,31.5%,5.4%,25.2%,17.9%,7.5%,20.7%。可见本文所提算法融合的图像对比度较高,可以同时更有效保留可见光图像的细节纹理和红外图像的特征信息,在同类方法中占有明显优势。展开更多
Epiduroscopy (EDS) allows diagnosis and treatment of spinal nerve root pathology by a fiberoptic device inserted percutaneously in the epidural space. Transient global amnesia syndrome (TGA) is defined by an abrupt on...Epiduroscopy (EDS) allows diagnosis and treatment of spinal nerve root pathology by a fiberoptic device inserted percutaneously in the epidural space. Transient global amnesia syndrome (TGA) is defined by an abrupt onset of an anterograde and retrograde amnesia that lasts up to 24 h, although mild subclinical neuropsychological deficits, with concomitant vegetative symptoms, can last for days after the episode. To our knowledge no cases of TGA in association with EDS have been recognized. We present a case of a 54-year-old man who was admitted for EDS, having experienced two times the failed surgery for spinal and left leg pain. During the EDS, 10 ml of non-ionic contrast media (CM) was injected to confirm the inside lesions. The epidural adhesions were gently broken down and drugs were injected around any inflamed nerve roots. Immediately after EDS, the patient experienced a TGA with loss of memory for the recent procedure and an impaired ability to retain new information. There were no signs of confusion or altered conscious level or seizure activities and neurological examination was normal. At 24 h a cranial computerized tomography (cCT) without CM showed “abnormal contrast enhancement at subarachnoid liquoral spaces levels”, then at 72 h cCT control was normal. The patient was discharged on the seventh day with perioperative lacunar amnesia persisting at this time. Herein, we sought a possible relationship between neurotoxic effect of CM and TGA in a patient emotionally stressed, even if the CM dose was given as a normal clinical range.展开更多
针对现有显著性检测模型准确度不高的问题,提出一种应用局部特征和全局特征对比的显著性检测模型.该算法首先使用简单的线性迭代聚类(Simple Linear Iterative Clustering,SLIC)分割算法将图像预分割为若干紧凑的超像素,选取边界区域集...针对现有显著性检测模型准确度不高的问题,提出一种应用局部特征和全局特征对比的显著性检测模型.该算法首先使用简单的线性迭代聚类(Simple Linear Iterative Clustering,SLIC)分割算法将图像预分割为若干紧凑的超像素,选取边界区域集并计算所有超像素的边界权重;然后计算颜色和纹理特征的局部对比度得到局部显著图,利用全局特征的独特性,空间分布特性得到全局显著图;最后采用求和乘积(Sum and Product,SP)方法将局部和全局显著图融合得到最终的显著图.在Achanta测试集上进行对比分析,实验结果表明本文算法能更准确地检测出显著区域,与其它5种算法相比具有较大的优势.展开更多
基金Supported by the China Postdoctoral Science Foundation(2011M500917)the Jiangsu Innovation Program for Graduate Education(CXLX11_0180)
文摘A new algorithm for bottom-up saliency estimation is proposed.Based on the sparse coding model,a power spectral filter is proposed to eliminate the second-order residual correlation,which suppresses the global repeated items effectively.In addition,aiming at modeling the mechanism of the human retina prior response to high-contrast stimuli,the effect of color context is considered.Experiments on the three publicly available databases and some psychophysical images show that the proposed model is comparable with the state-of-the-art saliency models,which not only highlights the salient objects in a complex environment but also pops up them uniformly.
基金the Natural Science Foundation of China(Nos.61602349,61375053,and 61273225)the China Scholarship Council(No.201508420248)Hubei Chengguang Talented Youth Development Foundation(No.2015B22)
文摘The goal of salient object detection is to estimate the regions which are most likely to attract human's visual attention. As an important image preprocessing procedure to reduce the computational complexity, salient object detection is still a challenging problem in computer vision. In this paper, we proposed a salient object detection model by integrating local and global superpixel contrast at multiple scales. Three features are computed to estimate the saliency of superpixel. Two optimization measures are utilized to refine the resulting saliency map. Extensive experiments with the state-of-the-art saliency models on four public datasets demonstrate the effectiveness of the proposed model.
文摘针对红外图像与可见光图像融合中细节丢失严重,红外图像的特征信息未能突出显示以及源图像的语义信息被忽视的问题,提出一种基于二次图像分解的红外图像与可见光图像融合网络(Secondary Image Decomposition For Infrared And Visible Image Fusion,SIDFuse)。利用编码器对源图像进行二次分解以提取不同尺度的特征信息,然后利用双元素注意力为不同尺度的特征信息分配权重、引入全局语义支路,再采用像素相加法作为融合策略,最后通过解码器重建融合图像。实验选择FLIR数据集用于训练,采用TNO和RoadScene两个数据集进行测试,并选取八种图像融合客观评价参数进行实验对比分析。由TNO数据集的图像融合实验表明,在信息熵、标准差、空间频率、视觉保真度、平均梯度、差异相关系数、多层级结构相似性、梯度融合性能评价指标上,SIDFuse比基于卷积网络中经典融合算法DenseFuse分别平均提高12.2%,9.0%,90.2%,13.9%,85.1%,16.8%,6.7%,30.7%,比最新的融合网络LRRNet分别平均提高2.5%,5.6%,31.5%,5.4%,25.2%,17.9%,7.5%,20.7%。可见本文所提算法融合的图像对比度较高,可以同时更有效保留可见光图像的细节纹理和红外图像的特征信息,在同类方法中占有明显优势。
文摘Epiduroscopy (EDS) allows diagnosis and treatment of spinal nerve root pathology by a fiberoptic device inserted percutaneously in the epidural space. Transient global amnesia syndrome (TGA) is defined by an abrupt onset of an anterograde and retrograde amnesia that lasts up to 24 h, although mild subclinical neuropsychological deficits, with concomitant vegetative symptoms, can last for days after the episode. To our knowledge no cases of TGA in association with EDS have been recognized. We present a case of a 54-year-old man who was admitted for EDS, having experienced two times the failed surgery for spinal and left leg pain. During the EDS, 10 ml of non-ionic contrast media (CM) was injected to confirm the inside lesions. The epidural adhesions were gently broken down and drugs were injected around any inflamed nerve roots. Immediately after EDS, the patient experienced a TGA with loss of memory for the recent procedure and an impaired ability to retain new information. There were no signs of confusion or altered conscious level or seizure activities and neurological examination was normal. At 24 h a cranial computerized tomography (cCT) without CM showed “abnormal contrast enhancement at subarachnoid liquoral spaces levels”, then at 72 h cCT control was normal. The patient was discharged on the seventh day with perioperative lacunar amnesia persisting at this time. Herein, we sought a possible relationship between neurotoxic effect of CM and TGA in a patient emotionally stressed, even if the CM dose was given as a normal clinical range.
文摘针对现有显著性检测模型准确度不高的问题,提出一种应用局部特征和全局特征对比的显著性检测模型.该算法首先使用简单的线性迭代聚类(Simple Linear Iterative Clustering,SLIC)分割算法将图像预分割为若干紧凑的超像素,选取边界区域集并计算所有超像素的边界权重;然后计算颜色和纹理特征的局部对比度得到局部显著图,利用全局特征的独特性,空间分布特性得到全局显著图;最后采用求和乘积(Sum and Product,SP)方法将局部和全局显著图融合得到最终的显著图.在Achanta测试集上进行对比分析,实验结果表明本文算法能更准确地检测出显著区域,与其它5种算法相比具有较大的优势.