To overcome the shortcomings of the Lee image enhancement algorithm and its improvement based on the logarithmic image processing(LIP) model, this paper proposes what we believe to be an effective image enhancement al...To overcome the shortcomings of the Lee image enhancement algorithm and its improvement based on the logarithmic image processing(LIP) model, this paper proposes what we believe to be an effective image enhancement algorithm. This algorithm introduces fuzzy entropy, makes full use of neighborhood information, fuzzy information and human visual characteristics.To enhance an image, this paper first carries out the reasonable fuzzy-3 partition of its histogram into the dark region, intermediate region and bright region. It then extracts the statistical characteristics of the three regions and adaptively selects the parameter αaccording to the statistical characteristics of the image’s gray-scale values. It also adds a useful nonlinear transform, thus increasing the ubiquity of the algorithm. Finally, the causes for the gray-scale value overcorrection that occurs in the traditional image enhancement algorithms are analyzed and their solutions are proposed.The simulation results show that our image enhancement algorithm can effectively suppress the noise of an image, enhance its contrast and visual effect, sharpen its edge and adjust its dynamic range.展开更多
在甲状腺癌手术决策中,难以在术前准确评估淋巴结转移。为减少不必要的手术,提升患者生活质量,准确预测甲状腺癌淋巴结转移具有重要意义。文中综合应用非负潜在因子模型和高效参数微调(Parameter Efficient Fine Tuning,PEFT)技术解决...在甲状腺癌手术决策中,难以在术前准确评估淋巴结转移。为减少不必要的手术,提升患者生活质量,准确预测甲状腺癌淋巴结转移具有重要意义。文中综合应用非负潜在因子模型和高效参数微调(Parameter Efficient Fine Tuning,PEFT)技术解决医学数据规模小和临床数据缺失问题。通过非负潜在因子模型对临床数据进行补全,提高数据可靠性和准确性。通过引入PEFT技术微调大型预训练模型,减少了计算成本。结果表明,在不同缺失比例下,潜在因子模型相较于传统方法更优越,PEFT方法在两个不同数据集上的训练精度较高且降低了训练时间。通过本地数据集和公开数据集的综合性能比较验证了所提方法的有效性。所提方法在保持高预测精度的同时降低了计算成本,具备更高的可解释性,为预训练大模型在医学任务中的应用提供了高效可行的方案。展开更多
Reconstructing the shape of a bubble will lay a firm foundation for further description of the dynamic characteristics of bubbly flow, especially for a single rising bubble or separate bubbles whose interaction could ...Reconstructing the shape of a bubble will lay a firm foundation for further description of the dynamic characteristics of bubbly flow, especially for a single rising bubble or separate bubbles whose interaction could be neglected. In this case, the rising bubble is usually simulated as an ellipsoid consisting of two semi-eUipsoids up and down. Thus the projected image of a bubble consists of two semi-ellipses. In this paper, a method for reconstructing the ellipsoid bubble model is described following digital image processing, using the Hough transform in 2D ellipse parameter extraction which could cover most of the bubble edge points in the image. Then a method based on characteristic symmetric matrix is described to detect 3D bubble ellipsoid model parameters from 2D ellipse parameters of projection planes. This method can be applied to bubbles rising with low-velocity in static flow field much in conformity with the projection theory and the shape variation of the rising bubble. This method does not need to solve nonlinear equation sets and provides an easy way to calculate the characteristic matrix of a space ellipsoid model for deformed bubble. For bubble application, two assumed conditions and a calibration factor are proposed to simplify calculation and detection. Errors of ellipsoid center and three axes are minor. Errors of the three rotation angles have no negative effect on further study on bubbly flow.展开更多
基金supported by the National Natural Science Foundation of China(61472324)
文摘To overcome the shortcomings of the Lee image enhancement algorithm and its improvement based on the logarithmic image processing(LIP) model, this paper proposes what we believe to be an effective image enhancement algorithm. This algorithm introduces fuzzy entropy, makes full use of neighborhood information, fuzzy information and human visual characteristics.To enhance an image, this paper first carries out the reasonable fuzzy-3 partition of its histogram into the dark region, intermediate region and bright region. It then extracts the statistical characteristics of the three regions and adaptively selects the parameter αaccording to the statistical characteristics of the image’s gray-scale values. It also adds a useful nonlinear transform, thus increasing the ubiquity of the algorithm. Finally, the causes for the gray-scale value overcorrection that occurs in the traditional image enhancement algorithms are analyzed and their solutions are proposed.The simulation results show that our image enhancement algorithm can effectively suppress the noise of an image, enhance its contrast and visual effect, sharpen its edge and adjust its dynamic range.
文摘在甲状腺癌手术决策中,难以在术前准确评估淋巴结转移。为减少不必要的手术,提升患者生活质量,准确预测甲状腺癌淋巴结转移具有重要意义。文中综合应用非负潜在因子模型和高效参数微调(Parameter Efficient Fine Tuning,PEFT)技术解决医学数据规模小和临床数据缺失问题。通过非负潜在因子模型对临床数据进行补全,提高数据可靠性和准确性。通过引入PEFT技术微调大型预训练模型,减少了计算成本。结果表明,在不同缺失比例下,潜在因子模型相较于传统方法更优越,PEFT方法在两个不同数据集上的训练精度较高且降低了训练时间。通过本地数据集和公开数据集的综合性能比较验证了所提方法的有效性。所提方法在保持高预测精度的同时降低了计算成本,具备更高的可解释性,为预训练大模型在医学任务中的应用提供了高效可行的方案。
基金support from the National Natural Science Foundation of China(No.51176141)the Natural Science Foundation of Tianjin(No.11JCZDJC22500)
文摘Reconstructing the shape of a bubble will lay a firm foundation for further description of the dynamic characteristics of bubbly flow, especially for a single rising bubble or separate bubbles whose interaction could be neglected. In this case, the rising bubble is usually simulated as an ellipsoid consisting of two semi-eUipsoids up and down. Thus the projected image of a bubble consists of two semi-ellipses. In this paper, a method for reconstructing the ellipsoid bubble model is described following digital image processing, using the Hough transform in 2D ellipse parameter extraction which could cover most of the bubble edge points in the image. Then a method based on characteristic symmetric matrix is described to detect 3D bubble ellipsoid model parameters from 2D ellipse parameters of projection planes. This method can be applied to bubbles rising with low-velocity in static flow field much in conformity with the projection theory and the shape variation of the rising bubble. This method does not need to solve nonlinear equation sets and provides an easy way to calculate the characteristic matrix of a space ellipsoid model for deformed bubble. For bubble application, two assumed conditions and a calibration factor are proposed to simplify calculation and detection. Errors of ellipsoid center and three axes are minor. Errors of the three rotation angles have no negative effect on further study on bubbly flow.
