Aiming at the problem of music noise introduced by classical spectral subtraction,a shorttime modulation domain(STM)spectral subtraction method has been successfully applied for singlechannel speech enhancement.Howeve...Aiming at the problem of music noise introduced by classical spectral subtraction,a shorttime modulation domain(STM)spectral subtraction method has been successfully applied for singlechannel speech enhancement.However,due to the inaccurate voice activity detection(VAD),the residual music noise and enhanced performance still need to be further improved,especially in the low signal to noise ratio(SNR)scenarios.To address this issue,an improved frame iterative spectral subtraction in the STM domain(IMModSSub)is proposed.More specifically,with the inter-frame correlation,the noise subtraction is directly applied to handle the noisy signal for each frame in the STM domain.Then,the noisy signal is classified into speech or silence frames based on a predefined threshold of segmented SNR.With these classification results,a corresponding mask function is developed for noisy speech after noise subtraction.Finally,exploiting the increased sparsity of speech signal in the modulation domain,the orthogonal matching pursuit(OMP)technique is employed to the speech frames for improving the speech quality and intelligibility.The effectiveness of the proposed method is evaluated with three types of noise,including white noise,pink noise,and hfchannel noise.The obtained results show that the proposed method outperforms some established baselines at lower SNRs(-5 to +5 dB).展开更多
A fast image segmentation algorithm based on salient features model and spatial-frequency domain adaptive kernel is proposed to solve the accurate discriminate objects problem of online visual detection in such scenes...A fast image segmentation algorithm based on salient features model and spatial-frequency domain adaptive kernel is proposed to solve the accurate discriminate objects problem of online visual detection in such scenes of variable sample morphological characteristics,low contrast and complex background texture.Firstly,by analyzing the spectral component distribution and spatial contour feature of the image,a salient feature model is established in spatial-frequency domain.Then,the salient object detection method based on Gaussian band-pass filter and the design criterion of adaptive convolution kernel are proposed to extract the salient contour feature of the target in spatial and frequency domain.Finally,the selection and growth rules of seed points are improved by integrating the gray level and contour features of the target,and the target is segmented by seeded region growing.Experiments have been performed on Berkeley Segmentation Data Set,as well as sample images of online detection,to verify the effectiveness of the algorithm.The experimental results show that the Jaccard Similarity Coefficient of the segmentation is more than 90%,which indicates that the proposed algorithm can availably extract the target feature information,suppress the background texture and resist noise interference.Besides,the Hausdorff Distance of the segmentation is less than 10,which infers that the proposed algorithm obtains a high evaluation on the target contour preservation.The experimental results also show that the proposed algorithm significantly improves the operation efficiency while obtaining comparable segmentation performance over other algorithms.展开更多
Most noise suppression algorithms of single channel use the mean of noisy segments to estimate the characteristics of noise spectrum, ignoring the estimation of noise in speech segments. Therefore, when the energy lev...Most noise suppression algorithms of single channel use the mean of noisy segments to estimate the characteristics of noise spectrum, ignoring the estimation of noise in speech segments. Therefore, when the energy level of noise varies with the time, the performance of removing noise will be degraded. To solve this problem, a speech enhancement approach based on dynamic noise estimation within correlation domain was proposed. This method exploits the characteristics that noise energy mainly concentrates on 0 th order correlation coefficients, signal is auto correlated but signal and noise, noise and noise are uncorrelated, then estimates and decomposes the noise, thus helps to solve the above mentioned problem. The results of recognition experiments on speech signals of 15 Chinese cities’ names corrupted by noise of exhibition hall shows, this approach is better than SS (Spectral Subtraction) method, adapts better to the variances of energy levels of speech signal corrupted by noise, has some practicability to improve the robustness of recognition systems under noisy environment.展开更多
In this paper, modifications to the finite-difference time-domain(FD-TD) method for modeling microwave pulse coupling into a slot, which is much narrower than one conventional FD-TD cell, are discussed. The coupling p...In this paper, modifications to the finite-difference time-domain(FD-TD) method for modeling microwave pulse coupling into a slot, which is much narrower than one conventional FD-TD cell, are discussed. The coupling process of microwave pulse into a slot is studied by using the modified FD-TD method, and the dependence of microwave coupling on slot sizes, the carrier frequencies and the polarization directions of the incident waves is analysed. Resonant and enhancement effects which occur in this process are observed. The condition at which the resonant effect takes place is also presented.展开更多
目的:基于文献计量学探讨国内外加速康复外科(enhanced recovery after surgery,ERAS)领域的研究热点及发展趋势。方法:利用中国国家知识基础设施(China National Knowledge Infrastructure,CNKI)检索平台、万方数据知识服务平台和Web o...目的:基于文献计量学探讨国内外加速康复外科(enhanced recovery after surgery,ERAS)领域的研究热点及发展趋势。方法:利用中国国家知识基础设施(China National Knowledge Infrastructure,CNKI)检索平台、万方数据知识服务平台和Web of Science核心合集检索相关中外文文献,检索时间范围为建库至2023年2月10日,使用Citespace 6.1.R6、VOSviewer 1.6.13软件可视化分析年发文量、期刊、作者、国家、机构、学科和关键词信息。结果:共纳入中文6166篇、外文6212篇。ERAS相关中外文献的年发文量均呈逐年上升趋势;《齐鲁护理杂志》和Surgical Endoscopy and Other Interventional Techniques分别为国内外发文量最多的期刊,前10位共被引期刊的2022年期刊影响因子(Journal Impact Factor,JIF)多高于5分;美国、中国、英国的外文发文量排名前三,且差距明显,美国对外合作最为广泛;四川大学华西医院和哥本哈根大学发文量最多,伦敦大学学院是对外合作最密切的科研机构;国内外发文量最多的作者分别为江志伟和Kehlet H;学科主要分布在外科学、麻醉学、胃肠病学与肝病学等;中外文高频关键词为围手术期、腹腔镜、结直肠肿瘤、胃肠肿瘤、并发症、length of stay、colorectal surgery、perioperative care和postoperative complications等;研究热点集中于ERAS的外科应用、围手术期管理、应用效果和循证研究;达芬奇机器人、并发症、生活质量、预康复、循证护理、multimodal analgesia、pain management、same-day discharge等突现词体现了研究前沿。结论:国内ERAS的研究内容以临床/护理应用效果为主,同质性较高,创新性略显不足,缺乏大样本前瞻性研究,且实施过程存在困难和瓶颈,需重视高质量研究的开展,借鉴国外ERAS专业团队的成功经验,探索具有中国特色的ERAS发展之路。展开更多
原始采集的医学图像普遍存在对比度不足、细节模糊以及噪声干扰等质量问题,使得现有医学图像分割技术的精度很难达到新的突破。针对医学图像数据增强技术进行研究,在不明显改变图像外观的前提下,通过添加特定的像素补偿和进行细微的图...原始采集的医学图像普遍存在对比度不足、细节模糊以及噪声干扰等质量问题,使得现有医学图像分割技术的精度很难达到新的突破。针对医学图像数据增强技术进行研究,在不明显改变图像外观的前提下,通过添加特定的像素补偿和进行细微的图像调整来改善原始图像质量问题,从而提高图像分割准确率。首先,设计引入了一个新的优化器模块,以产生一个连续分布的空间作为迁移的目标域,该优化器模块接受数据集的标签作为输入,并将离散的标签数据映射到连续分布的医学图像中;其次,提出了一个基于对抗生成网络的EnGAN模型,并将优化器模块产生的迁移目标域用来指导对抗网络的目标生成,从而将改善的医学图像质量知识植入模型中实现图像增强。基于COVID-19数据集,实验中使用U-Net、U-Net+ResNet34、U-Net+Attn Res U-Net等卷积神经网络作为骨干网络,Dice系数和交并比分别达到了73.5%和69.3%、75.1%和70.5%,以及75.2%和70.3%。实验的结果表明,提出的医学图像质量增强技术在最大限度保留原始特征的条件下,有效地提高了分割的准确率,为后续的医学图像处理研究提供了一个更为稳健和高效的解决方案。展开更多
基于元学习的单源域泛化(single domain generalization,SDG)已成为解决领域偏移问题的有效技术之一。然而,源域和增强域的语义信息不一致以及域不变特征和域相关特征难以分离,使SDG模型难以实现良好的泛化性能。针对上述问题,提出了一...基于元学习的单源域泛化(single domain generalization,SDG)已成为解决领域偏移问题的有效技术之一。然而,源域和增强域的语义信息不一致以及域不变特征和域相关特征难以分离,使SDG模型难以实现良好的泛化性能。针对上述问题,提出了一种单源域泛化中基于域增强和特征对齐的元学习方案(meta-learning based on domain enhancement and feature alignment,MetaDefa)。利用背景替换和视觉损害技术为每一张图像生成多样且有效的增强图像,保证了源域和增强域之间的语义信息一致性;多通道特征对齐模块通过关注源域和增强域特征空间之间的相似目标区域和抑制非目标区域的特征表示充分挖掘图像信息,进而有效地提取充足的可迁移性知识。通过实验评估,MetaDefa在office-Caltech-10、office31和PACS数据集上分别取得了88.87%、73.06%和57.06%的精确度。结果表明,MetaDefa方法成功实现了源图像和增强图像之间的语义一致性和对域不变特征的充分提取,从而显著提升了单源域泛化模型的泛化性能。展开更多
基金National Natural Science Foundation of China(NSFC)(No.61671075)Major Program of National Natural Science Foundation of China(No.61631003)。
文摘Aiming at the problem of music noise introduced by classical spectral subtraction,a shorttime modulation domain(STM)spectral subtraction method has been successfully applied for singlechannel speech enhancement.However,due to the inaccurate voice activity detection(VAD),the residual music noise and enhanced performance still need to be further improved,especially in the low signal to noise ratio(SNR)scenarios.To address this issue,an improved frame iterative spectral subtraction in the STM domain(IMModSSub)is proposed.More specifically,with the inter-frame correlation,the noise subtraction is directly applied to handle the noisy signal for each frame in the STM domain.Then,the noisy signal is classified into speech or silence frames based on a predefined threshold of segmented SNR.With these classification results,a corresponding mask function is developed for noisy speech after noise subtraction.Finally,exploiting the increased sparsity of speech signal in the modulation domain,the orthogonal matching pursuit(OMP)technique is employed to the speech frames for improving the speech quality and intelligibility.The effectiveness of the proposed method is evaluated with three types of noise,including white noise,pink noise,and hfchannel noise.The obtained results show that the proposed method outperforms some established baselines at lower SNRs(-5 to +5 dB).
基金supported by National Natural Science Foundation of China[grant numbers 61573233]Natural Science Foundation of Guangdong,China[grant numbers 2021A1515010661]+1 种基金Special projects in key fields of colleges and universities in Guangdong Province[grant numbers 2020ZDZX2005]Innovation Team Project of University in Guangdong Province[grant numbers 2015KCXTD018].
文摘A fast image segmentation algorithm based on salient features model and spatial-frequency domain adaptive kernel is proposed to solve the accurate discriminate objects problem of online visual detection in such scenes of variable sample morphological characteristics,low contrast and complex background texture.Firstly,by analyzing the spectral component distribution and spatial contour feature of the image,a salient feature model is established in spatial-frequency domain.Then,the salient object detection method based on Gaussian band-pass filter and the design criterion of adaptive convolution kernel are proposed to extract the salient contour feature of the target in spatial and frequency domain.Finally,the selection and growth rules of seed points are improved by integrating the gray level and contour features of the target,and the target is segmented by seeded region growing.Experiments have been performed on Berkeley Segmentation Data Set,as well as sample images of online detection,to verify the effectiveness of the algorithm.The experimental results show that the Jaccard Similarity Coefficient of the segmentation is more than 90%,which indicates that the proposed algorithm can availably extract the target feature information,suppress the background texture and resist noise interference.Besides,the Hausdorff Distance of the segmentation is less than 10,which infers that the proposed algorithm obtains a high evaluation on the target contour preservation.The experimental results also show that the proposed algorithm significantly improves the operation efficiency while obtaining comparable segmentation performance over other algorithms.
文摘Most noise suppression algorithms of single channel use the mean of noisy segments to estimate the characteristics of noise spectrum, ignoring the estimation of noise in speech segments. Therefore, when the energy level of noise varies with the time, the performance of removing noise will be degraded. To solve this problem, a speech enhancement approach based on dynamic noise estimation within correlation domain was proposed. This method exploits the characteristics that noise energy mainly concentrates on 0 th order correlation coefficients, signal is auto correlated but signal and noise, noise and noise are uncorrelated, then estimates and decomposes the noise, thus helps to solve the above mentioned problem. The results of recognition experiments on speech signals of 15 Chinese cities’ names corrupted by noise of exhibition hall shows, this approach is better than SS (Spectral Subtraction) method, adapts better to the variances of energy levels of speech signal corrupted by noise, has some practicability to improve the robustness of recognition systems under noisy environment.
文摘In this paper, modifications to the finite-difference time-domain(FD-TD) method for modeling microwave pulse coupling into a slot, which is much narrower than one conventional FD-TD cell, are discussed. The coupling process of microwave pulse into a slot is studied by using the modified FD-TD method, and the dependence of microwave coupling on slot sizes, the carrier frequencies and the polarization directions of the incident waves is analysed. Resonant and enhancement effects which occur in this process are observed. The condition at which the resonant effect takes place is also presented.
文摘目的:基于文献计量学探讨国内外加速康复外科(enhanced recovery after surgery,ERAS)领域的研究热点及发展趋势。方法:利用中国国家知识基础设施(China National Knowledge Infrastructure,CNKI)检索平台、万方数据知识服务平台和Web of Science核心合集检索相关中外文文献,检索时间范围为建库至2023年2月10日,使用Citespace 6.1.R6、VOSviewer 1.6.13软件可视化分析年发文量、期刊、作者、国家、机构、学科和关键词信息。结果:共纳入中文6166篇、外文6212篇。ERAS相关中外文献的年发文量均呈逐年上升趋势;《齐鲁护理杂志》和Surgical Endoscopy and Other Interventional Techniques分别为国内外发文量最多的期刊,前10位共被引期刊的2022年期刊影响因子(Journal Impact Factor,JIF)多高于5分;美国、中国、英国的外文发文量排名前三,且差距明显,美国对外合作最为广泛;四川大学华西医院和哥本哈根大学发文量最多,伦敦大学学院是对外合作最密切的科研机构;国内外发文量最多的作者分别为江志伟和Kehlet H;学科主要分布在外科学、麻醉学、胃肠病学与肝病学等;中外文高频关键词为围手术期、腹腔镜、结直肠肿瘤、胃肠肿瘤、并发症、length of stay、colorectal surgery、perioperative care和postoperative complications等;研究热点集中于ERAS的外科应用、围手术期管理、应用效果和循证研究;达芬奇机器人、并发症、生活质量、预康复、循证护理、multimodal analgesia、pain management、same-day discharge等突现词体现了研究前沿。结论:国内ERAS的研究内容以临床/护理应用效果为主,同质性较高,创新性略显不足,缺乏大样本前瞻性研究,且实施过程存在困难和瓶颈,需重视高质量研究的开展,借鉴国外ERAS专业团队的成功经验,探索具有中国特色的ERAS发展之路。
文摘原始采集的医学图像普遍存在对比度不足、细节模糊以及噪声干扰等质量问题,使得现有医学图像分割技术的精度很难达到新的突破。针对医学图像数据增强技术进行研究,在不明显改变图像外观的前提下,通过添加特定的像素补偿和进行细微的图像调整来改善原始图像质量问题,从而提高图像分割准确率。首先,设计引入了一个新的优化器模块,以产生一个连续分布的空间作为迁移的目标域,该优化器模块接受数据集的标签作为输入,并将离散的标签数据映射到连续分布的医学图像中;其次,提出了一个基于对抗生成网络的EnGAN模型,并将优化器模块产生的迁移目标域用来指导对抗网络的目标生成,从而将改善的医学图像质量知识植入模型中实现图像增强。基于COVID-19数据集,实验中使用U-Net、U-Net+ResNet34、U-Net+Attn Res U-Net等卷积神经网络作为骨干网络,Dice系数和交并比分别达到了73.5%和69.3%、75.1%和70.5%,以及75.2%和70.3%。实验的结果表明,提出的医学图像质量增强技术在最大限度保留原始特征的条件下,有效地提高了分割的准确率,为后续的医学图像处理研究提供了一个更为稳健和高效的解决方案。
文摘基于元学习的单源域泛化(single domain generalization,SDG)已成为解决领域偏移问题的有效技术之一。然而,源域和增强域的语义信息不一致以及域不变特征和域相关特征难以分离,使SDG模型难以实现良好的泛化性能。针对上述问题,提出了一种单源域泛化中基于域增强和特征对齐的元学习方案(meta-learning based on domain enhancement and feature alignment,MetaDefa)。利用背景替换和视觉损害技术为每一张图像生成多样且有效的增强图像,保证了源域和增强域之间的语义信息一致性;多通道特征对齐模块通过关注源域和增强域特征空间之间的相似目标区域和抑制非目标区域的特征表示充分挖掘图像信息,进而有效地提取充足的可迁移性知识。通过实验评估,MetaDefa在office-Caltech-10、office31和PACS数据集上分别取得了88.87%、73.06%和57.06%的精确度。结果表明,MetaDefa方法成功实现了源图像和增强图像之间的语义一致性和对域不变特征的充分提取,从而显著提升了单源域泛化模型的泛化性能。