为了提高复杂背景下红外小目标的检测能力,提出了一种改进加权增强局部对比算法(weighted strengthened local contrast measure,WSLCM)的红外小目标检测方法。在WSLCM的预处理阶段采用自适应曲率滤波在多尺度下对图像进行处理,抑制背...为了提高复杂背景下红外小目标的检测能力,提出了一种改进加权增强局部对比算法(weighted strengthened local contrast measure,WSLCM)的红外小目标检测方法。在WSLCM的预处理阶段采用自适应曲率滤波在多尺度下对图像进行处理,抑制背景过程中保持真实目标不被淹没。在背景抑制计算中选取背景块中的灰度最大值作为背景估计,降低虚警率。同时引入目标增强因子与背景抑制因子,加强算法鲁棒性,消除背景噪声影响,增强对红外小目标的检测能力;通过嵌入式ZYNQ平台进行算法IP核定制,采用软硬件协同方式实现算法对特定场景下小目标的识别检测。实验表明,相比于传统WSLCM算法BSF和SCRG指标都有明显提高,连续帧检测率为93.2%,嵌入式平台检测效率比PC端提升17.6%,验证了算法和嵌入式系统的有效性。展开更多
An integrated approach to the study of the fault patterns and seismic attributes was carried out on the OT field using a 3D swamp seismic data covering approximately 420 Km2 of western belt of the Niger Delta. This st...An integrated approach to the study of the fault patterns and seismic attributes was carried out on the OT field using a 3D swamp seismic data covering approximately 420 Km2 of western belt of the Niger Delta. This study aimed at improving the visualization of faults in the study area using different seismic attributes. Three major growth faults dipping south with few antithetic faults dipping in opposite direction to the growth faults and other minor faults were identified manually on the original and steered seismic cube in the inline 11669, while in the cross line only one major fault was identified manually. The seismic volume was subjected to several stages of post-stack processing to enhance discontinuities. First, a dip-steering volume was created. Several dip-steered filters were then applied to enhance faults and fractures visualization. Finally, similarity and curvature attributes were calculated on the dip-steered and fault enhanced volume. These final attributes show detailed geometry of the fault system and the numerous subtle lineaments in the study area. The integration of the attributes has increased confidence in the seismic mapping of the faults and the other numerous subtle lineaments which were difficult to identify on the input data. Similarity and curvature attributes of the seismic volume preserve subtle structural details and permit a more robust interpretation of the structures.展开更多
韦伯局部描述子(Weber Local Descriptor,WLD)被广泛用在多个领域,但由于其所具有的局限性使它对图像中的位移和旋转不具有较好的鲁棒性。因此,本文对WLD进行改进,首先,提出一种多尺度可变曲率Gabor滤波器对指静脉图像进行滤波,得到多...韦伯局部描述子(Weber Local Descriptor,WLD)被广泛用在多个领域,但由于其所具有的局限性使它对图像中的位移和旋转不具有较好的鲁棒性。因此,本文对WLD进行改进,首先,提出一种多尺度可变曲率Gabor滤波器对指静脉图像进行滤波,得到多尺度能量图和几何特征图;然后,用能量图计算差分激励,在计算时采用多尺度局部窗口,并加入方向信息;最后将几何特征图和差分激励结合形成特征。该方法的有效性将在FV-TJ和FV-USM数据库上进行验证,结果表明本文方法的识别性能要优于其他方法。展开更多
文摘为了提高复杂背景下红外小目标的检测能力,提出了一种改进加权增强局部对比算法(weighted strengthened local contrast measure,WSLCM)的红外小目标检测方法。在WSLCM的预处理阶段采用自适应曲率滤波在多尺度下对图像进行处理,抑制背景过程中保持真实目标不被淹没。在背景抑制计算中选取背景块中的灰度最大值作为背景估计,降低虚警率。同时引入目标增强因子与背景抑制因子,加强算法鲁棒性,消除背景噪声影响,增强对红外小目标的检测能力;通过嵌入式ZYNQ平台进行算法IP核定制,采用软硬件协同方式实现算法对特定场景下小目标的识别检测。实验表明,相比于传统WSLCM算法BSF和SCRG指标都有明显提高,连续帧检测率为93.2%,嵌入式平台检测效率比PC端提升17.6%,验证了算法和嵌入式系统的有效性。
文摘An integrated approach to the study of the fault patterns and seismic attributes was carried out on the OT field using a 3D swamp seismic data covering approximately 420 Km2 of western belt of the Niger Delta. This study aimed at improving the visualization of faults in the study area using different seismic attributes. Three major growth faults dipping south with few antithetic faults dipping in opposite direction to the growth faults and other minor faults were identified manually on the original and steered seismic cube in the inline 11669, while in the cross line only one major fault was identified manually. The seismic volume was subjected to several stages of post-stack processing to enhance discontinuities. First, a dip-steering volume was created. Several dip-steered filters were then applied to enhance faults and fractures visualization. Finally, similarity and curvature attributes were calculated on the dip-steered and fault enhanced volume. These final attributes show detailed geometry of the fault system and the numerous subtle lineaments in the study area. The integration of the attributes has increased confidence in the seismic mapping of the faults and the other numerous subtle lineaments which were difficult to identify on the input data. Similarity and curvature attributes of the seismic volume preserve subtle structural details and permit a more robust interpretation of the structures.
文摘韦伯局部描述子(Weber Local Descriptor,WLD)被广泛用在多个领域,但由于其所具有的局限性使它对图像中的位移和旋转不具有较好的鲁棒性。因此,本文对WLD进行改进,首先,提出一种多尺度可变曲率Gabor滤波器对指静脉图像进行滤波,得到多尺度能量图和几何特征图;然后,用能量图计算差分激励,在计算时采用多尺度局部窗口,并加入方向信息;最后将几何特征图和差分激励结合形成特征。该方法的有效性将在FV-TJ和FV-USM数据库上进行验证,结果表明本文方法的识别性能要优于其他方法。
