CN114638766A - Method for correcting luminous remote sensing image - Google Patents
Method for correcting luminous remote sensing image Download PDFInfo
- Publication number
- CN114638766A CN114638766A CN202210364692.2A CN202210364692A CN114638766A CN 114638766 A CN114638766 A CN 114638766A CN 202210364692 A CN202210364692 A CN 202210364692A CN 114638766 A CN114638766 A CN 114638766A
- Authority
- CN
- China
- Prior art keywords
- remote sensing
- image
- luminous
- sensing image
- original
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 36
- 230000009466 transformation Effects 0.000 claims abstract description 22
- PXFBZOLANLWPMH-UHFFFAOYSA-N 16-Epiaffinine Natural products C1C(C2=CC=CC=C2N2)=C2C(=O)CC2C(=CC)CN(C)C1C2CO PXFBZOLANLWPMH-UHFFFAOYSA-N 0.000 claims abstract description 9
- 238000013519 translation Methods 0.000 claims description 28
- 238000012545 processing Methods 0.000 claims description 20
- 238000010586 diagram Methods 0.000 claims description 10
- 238000012800 visualization Methods 0.000 claims description 9
- 230000000007 visual effect Effects 0.000 claims description 6
- 238000004422 calculation algorithm Methods 0.000 claims description 4
- 238000003702 image correction Methods 0.000 abstract description 14
- 238000012937 correction Methods 0.000 abstract description 6
- 230000003287 optical effect Effects 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 3
- 238000000605 extraction Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 238000012952 Resampling Methods 0.000 description 1
- 238000009825 accumulation Methods 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 230000004298 light response Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000012732 spatial analysis Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/80—Geometric correction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/02—Affine transformations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10032—Satellite or aerial image; Remote sensing
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
Abstract
本发明提供一种夜光遥感影像校正方法，包括：基于初始夜光遥感影像确定目标区域，获取目标区域的道路矢量图；将道路矢量图由矢量数据转换为栅格数据，作为原始夜光遥感底图，栅格数据与初始夜光遥感影像的图像数据分辨率相同；获取夜光遥感影像，对夜光遥感影像进行可视化增强处理；基于夜光遥感影像在原始夜光遥感底图上确定控制点，其中，控制点用于表征夜光遥感影像在原始夜光遥感底图上的对应位置；基于控制点，利用仿射变换模型，对夜光遥感影像进行校正。本发明以道路矢量为基础制作原始夜光遥感校正底图，利用获取的夜光遥感影像数据与原始夜光遥感校正底图进行匹配比对，校正夜光遥感影像的数据，进而实现夜光遥感数据的高精度定位。
The invention provides a luminous remote sensing image correction method, comprising: determining a target area based on an initial luminous remote sensing image, and obtaining a road vector map of the target area; converting the road vector data from vector data to raster data, as the original luminous remote sensing base map, The resolution of the raster data is the same as that of the original luminous remote sensing image; the luminous remote sensing image is obtained, and the luminous remote sensing image is visually enhanced; Characterize the corresponding position of luminous remote sensing images on the original luminous remote sensing base map; based on control points, the affine transformation model is used to correct the luminous remote sensing images. The invention makes the original night light remote sensing correction base map based on the road vector, uses the acquired night light remote sensing image data and the original night light remote sensing correction base map to match and compare, and corrects the data of the night light remote sensing image, thereby realizing the high-precision positioning of the night light remote sensing data. .
Description
技术领域technical field
本发明涉及夜光遥感技术领域，尤其涉及一种夜光遥感影像校正方法。The invention relates to the technical field of night light remote sensing, in particular to a night light remote sensing image correction method.
背景技术Background technique
夜光遥感手段可以获取夜间灯光、火光等信息，拓展了光学遥感获取地表信息的时间宽度和信息维度，常用于城市发展评估、人类活动预测、经济指标分析、夜间渔船作业监管等领域。相对于日间光学遥感，夜光遥感仅能获取夜间亮度信息，主要包括灯光和火光等，其包含的影像纹理丰富度较低，现有的日间遥感技术很难应用于夜光遥感影像。Night light remote sensing can obtain information such as night lights and firelights, which expands the time width and information dimension of surface information obtained by optical remote sensing. Compared with daytime optical remote sensing, nighttime light remote sensing can only obtain nighttime brightness information, mainly including lights and firelight, etc., which contain low image texture richness, and the existing daytime remote sensing technology is difficult to apply to nighttime light remote sensing images.
现有的夜光遥感影技术中没有针对夜光遥感影像高精度定位的处理方法，也没有建立适用于夜光遥感影像的基础控制底图数据，往往采用日间光学遥感底图和人工标记方法进行控制信息提取，在大规模数据处理应用中无法保证夜光遥感数据的处理效率和精度。In the existing night light remote sensing image technology, there is no processing method for high-precision positioning of night light remote sensing images, and no basic control base map data suitable for night light remote sensing images has been established. Daytime optical remote sensing base maps and manual marking methods are often used to control information. Extraction, the processing efficiency and accuracy of night light remote sensing data cannot be guaranteed in large-scale data processing applications.
发明内容SUMMARY OF THE INVENTION
针对上述问题，本发明提出一种夜光遥感影像校正方法，以道路矢量为基础制作原始夜光遥感校正底图，利用获取的夜光遥感影像数据与原始夜光遥感校正底图进行匹配比对，校正夜光遥感影像的数据，进而实现夜光遥感数据的高精度定位。In view of the above problems, the present invention proposes a luminous remote sensing image correction method. Based on the road vector, an original luminous remote sensing correction base map is made, and the acquired luminous remote sensing image data is matched and compared with the original luminous remote sensing correction base map, and the luminous remote sensing is corrected. Image data, and then achieve high-precision positioning of night light remote sensing data.
本发明提出的夜光遥感影像校正方法包括：基于初始夜光遥感影像确定目标区域，获取目标区域的道路矢量图；将道路矢量图由矢量数据转换为栅格数据，作为原始夜光遥感底图，其中，栅格数据与初始夜光遥感影像的图像数据分辨率相同；获取夜光遥感影像，对夜光遥感影像进行可视化增强处理；基于夜光遥感影像在原始夜光遥感底图上确定控制点，其中，控制点用于表征夜光遥感影像在原始夜光遥感底图上的对应位置；基于控制点，利用仿射变换模型，对夜光遥感影像进行校正。The luminous remote sensing image correction method provided by the present invention includes: determining a target area based on an initial luminous remote sensing image, and obtaining a road vector map of the target area; converting the road vector data from vector data to raster data as the original luminous remote sensing base map, wherein, The resolution of the raster data is the same as that of the original night light remote sensing image; the night light remote sensing image is obtained, and the night light remote sensing image is visually enhanced; based on the night light remote sensing image, the control point is determined on the original night light remote sensing base map. Characterize the corresponding position of luminous remote sensing images on the original luminous remote sensing base map; based on control points, the affine transformation model is used to correct the luminous remote sensing images.
进一步地，本发明的夜光遥感影像校正方法中，基于夜光遥感影像在原始夜光遥感底图上确定控制点包括：将夜光遥感影像划分为多个区块，在多个区块中选取对比度大于第一阈值的区块，作为特征匹配块；将特征匹配块与原始夜光遥感底图进行模板匹配，确定所述特征匹配块相对于所述原始夜光遥感底图的第一平移量；选取第一平移量在第一范围内的原始夜光遥感底图的图像点作为夜光遥感影像的控制点。Further, in the luminous remote sensing image correction method of the present invention, determining the control point on the original luminous remote sensing base map based on the luminous remote sensing image includes: dividing the luminous remote sensing image into a plurality of blocks, and selecting a contrast ratio greater than the number of blocks in the plurality of blocks. A threshold block is used as a feature matching block; template matching is performed between the feature matching block and the original night light remote sensing base map, and the first translation amount of the feature matching block relative to the original night light remote sensing base map is determined; the first translation is selected The image points of the original luminous remote sensing base map with the amount within the first range are used as the control points of the luminous remote sensing image.
进一步地，本发明的夜光遥感影像校正方法中，确定特征匹配块相对于原始夜光遥感底图的第一平移量包括：计算特征匹配块与原始夜光遥感底图在各平移量下的图像相似度系数；将图像相似度系数最大时对应的平移量确定为第一平移量；其中，计算特征匹配块与原始夜光遥感底图在各平移量下的图像相似度系数采用如下公式：Further, in the luminous remote sensing image correction method of the present invention, determining the first translation amount of the feature matching block relative to the original luminous remote sensing base map includes: calculating the image similarity between the feature matching block and the original luminous remote sensing base map under each translation amount. coefficient; the corresponding translation amount when the image similarity coefficient is the largest is determined as the first translation amount; wherein, the following formula is used to calculate the image similarity coefficient between the feature matching block and the original night light remote sensing base map under each translation amount:
S coef (u,v)为图像相似度系数，(x,y)为原始夜光遥感底图位置坐标，(u,v)为特征匹配块相对于原始夜光遥感底图在(x,y)处的平移量，C(x-u,y-v)为特征匹配块在(x-u,y-v)处的图像灰度值，D(x,y)为原始夜光遥感底图在(x,y)处的图像灰度值，μ 1为特征匹配块的灰度均值，μ 2为原始夜光遥感底图的灰度均值，计算公式为：S coef ( u , v ) is the image similarity coefficient, ( x , y ) is the position coordinates of the original night light remote sensing base map, ( u , v ) is the feature matching block relative to the original night light remote sensing base map at ( x , y ) , C( x - u , y - v ) is the image gray value of the feature matching block at ( x - u , y - v ), D( x , y ) is the original night light remote sensing base map at ( x , y ), μ 1 is the gray mean value of the feature matching block, μ 2 is the gray mean value of the original luminous remote sensing base map, and the calculation formula is:
D(x-u,y-v)为原始夜光遥感底图在(x-u,y-v)处的图像灰度值，N为原始夜光遥感底图的像素数量。D( x - u , y - v ) is the image gray value of the original night light remote sensing base map at ( x - u , y - v ), and N is the number of pixels of the original night light remote sensing base map.
进一步地，本发明的夜光遥感影像校正方法中，基于夜光遥感影像在原始夜光遥感底图上确定控制点包括：将第一平移量在第一范围内的原始夜光遥感底图的图像点标记为第一类点；采用RANSAC算法判断第一类点是否符合整体一致性；将符合整体一致性的第一类点作为夜光遥感影像的控制点。Further, in the luminous remote sensing image correction method of the present invention, determining the control point on the original luminous remote sensing base map based on the luminous remote sensing image includes: marking the image point of the original luminous remote sensing base map with the first translation amount within the first range as The first type of point; the RANSAC algorithm is used to judge whether the first type of point conforms to the overall consistency; the first type of point that conforms to the overall consistency is used as the control point of the luminous remote sensing image.
进一步地，本发明的夜光遥感影像校正方法包括：对夜光遥感影像进行校正后，利用已校正的夜光遥感影像更新夜光遥感底图。Further, the luminous remote sensing image correction method of the present invention includes: after correcting the luminous remote sensing image, using the corrected luminous remote sensing image to update the luminous remote sensing base map.
进一步地，本发明的夜光遥感影像校正方法中，利用已校正的夜光遥感影像更新夜光遥感底图包括：基于已校正的夜光遥感影像的图像亮度，调整夜光遥感底图的像素灰度值，其中，对于已校正的夜光遥感影像中无光亮反应的区域，在夜光遥感底图中对应位置按第一比例减小像素灰度值，对于已校正的夜光遥感影像中有光亮反应的区域，在夜光遥感底图中对应位置按第二比例增大像素灰度值，第二比例基于已校正的夜光遥感影像的图像像素灰度值总和调整。Further, in the luminous remote sensing image correction method of the present invention, using the corrected luminous remote sensing image to update the luminous remote sensing base map includes: adjusting the pixel gray value of the luminous remote sensing base map based on the image brightness of the corrected luminous remote sensing image, wherein , for the corrected luminous remote sensing image, the pixel gray value is reduced according to the first proportion at the corresponding position in the luminous remote sensing base map. The corresponding position in the remote sensing base image increases the pixel gray value according to a second ratio, and the second ratio is adjusted based on the sum of the image pixel gray value of the corrected luminous remote sensing image.
进一步地，本发明的夜光遥感影像校正方法中，将道路矢量图由矢量数据转换为栅格数据包括：在道路矢量图中，将有道路的位置像素灰度值设置为255，无道路的位置像素灰度值设置为100。Further, in the luminous remote sensing image correction method of the present invention, converting the road vector map from vector data to raster data includes: in the road vector map, setting the pixel gray value of the position with the road to 255, and the position without the road as 255. The pixel gray value is set to 100.
进一步地，本发明的夜光遥感影像校正方法中，对夜光遥感影像进行可视化增强处理包括：对夜光遥感影像使用对数变换进行拉伸处理；对夜光遥感影像进行0~255可视化范围的拉伸处理，以使夜光遥感影像由16bit数据转换为8bit数据。Further, in the luminous remote sensing image correction method of the present invention, performing the visualization enhancement processing on the luminous remote sensing image includes: using logarithmic transformation to stretch the luminous remote sensing image; , so that the luminous remote sensing image is converted from 16bit data to 8bit data.
进一步地，本发明的夜光遥感影像校正方法中，对夜光遥感影像使用对数变换进行拉伸处理使用的公式为：Further, in the luminous remote sensing image correction method of the present invention, the formula used for stretching the luminous remote sensing image using logarithmic transformation is:
其中， p为夜光遥感影像的原始图像灰度值，L为对夜光遥感影像使用对数变换进行拉伸处理后的图像灰度值。Among them, p is the original image gray value of the luminous remote sensing image, and L is the image gray value after stretching the luminous remote sensing image using logarithmic transformation.
进一步地，本发明的夜光遥感影像校正方法中，对夜光遥感影像进行0~255可视化范围的拉伸处理使用的公式为：Further, in the luminous remote sensing image correction method of the present invention, the formula used for stretching the luminous remote sensing image with a visualization range of 0 to 255 is:
其中， L为夜光遥感影像经过对数变换进行拉伸处理后的图像灰度值，V为对夜光遥感影像进行0~255可视化范围的拉伸处理后的图像灰度值。Among them, L is the gray value of the image after stretching the luminous remote sensing image through logarithmic transformation, and V is the gray value of the image after stretching the luminous remote sensing image with a visual range of 0 to 255.
本发明的夜光遥感影像校正方法具有如下有益效果：The luminous remote sensing image correction method of the present invention has the following beneficial effects:
（1）基于道路矢量图制作原始夜光遥感地图，道路矢量图全球覆盖且位置准确，而道路两侧路灯的灯光具有稳定性好、形状特征明显的特点，可以使沿道路排布的路灯成为夜光遥感影像中重要的可识别特征；(1) Create the original luminous remote sensing map based on the road vector map. The road vector map has global coverage and accurate location. The lights on both sides of the road have the characteristics of good stability and obvious shape characteristics, which can make the street lights arranged along the road become luminous Important identifiable features in remote sensing images;
（2）通过夜光影像可视化增强、高对比度分块提取以及模板匹配处理等步骤，可以实现夜光遥感影像的自动高精度定位处理；(2) Through the steps of luminous image visualization enhancement, high-contrast block extraction and template matching processing, automatic high-precision positioning processing of luminous remote sensing images can be realized;
（3）通过精确定位后夜光遥感影像数据迭代更新夜光遥感底图，可以逐渐优化夜光遥感底图数据，有利于夜光遥感影像的高精度定位处理。(3) By iteratively updating the night light remote sensing basemap data after accurate positioning, the night light remote sensing basemap data can be gradually optimized, which is conducive to the high-precision positioning processing of night light remote sensing images.
附图说明Description of drawings
图1是本发明的夜光遥感影像校正方法流程示意图；1 is a schematic flow chart of a method for calibrating luminous remote sensing images of the present invention;
图2是本发明基于夜光遥感影像在原始夜光遥感底图上确定控制点的方法流程示意图；2 is a schematic flowchart of the method for determining a control point on an original luminous remote sensing base map based on a luminous remote sensing image according to the present invention;
图3是本发明的一些实施例中实施夜光遥感影像校正方法的整体流程示意图。FIG. 3 is a schematic diagram of an overall flow of implementing a method for calibrating a night light remote sensing image in some embodiments of the present invention.
具体实施方式Detailed ways
为使本发明的目的、技术方案和优点更加清楚明白，以下结合具体实施例，并参照附图，对本发明进一步详细说明。In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to specific embodiments and accompanying drawings.
参见图1，本发明的夜光遥感影像校正方法包括如下步骤：Referring to Fig. 1, the method for calibrating a night light remote sensing image of the present invention comprises the following steps:
S101，基于初始夜光遥感影像确定目标区域，获取目标区域的道路矢量图。S101: Determine a target area based on an initial night light remote sensing image, and obtain a road vector diagram of the target area.
夜光遥感数据主要获取夜间灯光和火光等数据，一般来说，火光数据不具有固定位置的属性，而路灯的灯光具有稳定性好、形状特征明显的特点，因此可以作为夜光遥感影像的可识别特征。另一方面，路灯具有沿道路布设的特点，利用道路矢量图即可拟合夜光遥感影像的灯光分布情况。Night light remote sensing data mainly obtains data such as night light and fire light. Generally speaking, fire light data does not have the attribute of fixed position, while street light has the characteristics of good stability and obvious shape characteristics, so it can be used as a identifiable feature of night light remote sensing images. . On the other hand, street lights have the characteristics of being arranged along the road, and the light distribution of the night light remote sensing image can be fitted by using the road vector diagram.
基于此，本发明采用道路矢量图作为原始夜光遥感地图的依据。作为导航交通的重要依据，道路矢量数据具有开源、全球覆盖、位置准确的特点。本发明可以直接一次性获取全球的道路矢量图，在经过初始夜光遥感影像确定目标区域后，再确定目标区域的道路矢量图。Based on this, the present invention adopts the road vector map as the basis of the original night light remote sensing map. As an important basis for navigating traffic, road vector data has the characteristics of open source, global coverage and accurate location. The invention can directly obtain the global road vector diagram at one time, and after determining the target area through the initial night light remote sensing image, the road vector diagram of the target area is determined.
在本发明的一些实施例中，根据遥感影像自带的空间定位信息，获得夜光遥感数据的空间覆盖范围，记为经度范围[lon1~lon2]，纬度范围[lat1~lat2]。其次，确定经度拓展范围为LonT，纬度拓展范围为LatT，则可以确定目标区域的空间范围为经度[(lon1-LonT)~(lon2+LonT)]，纬度[(lat1-LatT)~(lat2+LatT)]。利用区域裁切方法，在道路矢量图中确定该目标区域的道路矢量图，即为夜光遥感影像对应的道路矢量图，记为D′。In some embodiments of the present invention, the spatial coverage of night light remote sensing data is obtained according to the spatial positioning information carried by the remote sensing image, which is recorded as the longitude range [lon 1 ~lon 2 ] and the latitude range [lat 1 ~lat 2 ]. Secondly, determine that the longitude extension range is Lon T , and the latitude extension range is Lat T , then the spatial range of the target area can be determined to be longitude [(lon 1 -Lon T )~(lon 2 +Lon T )], latitude [(lat 1 -Lat T )~(lat 2 +Lat T )]. Using the area cropping method, the road vector map of the target area is determined in the road vector map, that is, the road vector map corresponding to the night light remote sensing image, which is denoted as D'.
S102，将道路矢量图由矢量数据转换为栅格数据，作为原始夜光遥感底图，其中，栅格数据与初始夜光遥感影像的图像数据分辨率相同。S102: Convert the road vector map from vector data to raster data as the original luminous remote sensing base map, wherein the raster data has the same resolution as the image data of the initial luminous remote sensing image.
栅格数据结构简单，空间分析和地理现象的模拟均比较容易，有利于与遥感数据的匹配应用和分析，因此本发明将已确定的目标区域的道路矢量图转换为栅格数据。The grid data structure is simple, the spatial analysis and the simulation of geographical phenomena are relatively easy, and it is beneficial to the matching application and analysis with the remote sensing data.
根据本发明的一些实施例，目标区域初始夜光遥感影像的空间分辨率经度方向为Rlon，纬度方向为Rlat，那么道路矢量图D′对应的栅格数据的尺寸为：横向(lon2-lon1-2LonT)/ Rlon，纵向(lat2-lat1-2LatT)/ Rlat。从而可以得到一幅与初始夜光遥感影像同分辨率的栅格影像图，此栅格化影像图即可作为原始夜光遥感底图，记为D。According to some embodiments of the present invention, the spatial resolution of the initial luminous remote sensing image of the target area is R lon in the longitude direction and R lat in the latitude direction, then the size of the raster data corresponding to the road vector diagram D' is: horizontal (lon 2 - lon 1 -2Lon T )/ R lon , Longitudinal (lat 2 -lat 1 -2Lat T )/ R lat . In this way, a raster image with the same resolution as the original luminous remote sensing image can be obtained, and this rasterized image can be used as the original luminous remote sensing base map, denoted as D.
根据本发明的一些实施例，在道路矢量图中，将有道路的位置像素灰度值设置为255，无道路的位置像素灰度值设置为100，从而得到一幅与初始夜光遥感影像同分辨率的栅格影像图。在实际操作中，将有道路的位置像素灰度值设置为255，无道路的位置像素灰度值设置为100，可以使得有无道路的位置更容易区分。According to some embodiments of the present invention, in the road vector diagram, the pixel gray value of the locus with a road is set to 255, and the gray value of the locus of the road without the road is set to 100, so as to obtain a luminous remote sensing image with the same resolution as the initial luminous remote sensing image. rate raster image. In actual operation, the gray value of the locus with road is set to 255, and the gray value of the locus without road is set to 100, which can make it easier to distinguish the position with or without road.
S103，获取夜光遥感影像，对夜光遥感影像进行可视化增强处理。S103 , acquiring a luminous remote sensing image, and performing visualization enhancement processing on the luminous remote sensing image.
夜光遥感影像主要获取的是夜间灯光和火光等数据，相对于日间光学遥感影像，其灰度值大多处于低亮度区域，且灰度直方图存在明显的长尾分布，不同于常见的近高斯分布。Night light remote sensing images mainly acquire data such as night lights and firelights. Compared with daytime optical remote sensing images, most of the gray values are in the low-brightness area, and the gray histogram has an obvious long-tail distribution, which is different from the common near-Gaussian image. distributed.
同时，遥感影像普遍采用16bit数据存储，而计算显示和图像处理运算常在8bit数据中进行。因此，需要对夜光遥感影像进行可视化增强处理，并将夜光遥感影像转换为8bit数据。At the same time, remote sensing images generally use 16-bit data storage, while calculation display and image processing operations are often performed in 8-bit data. Therefore, it is necessary to visually enhance the remote sensing image of night light, and convert the remote sensing image of night light into 8bit data.
根据本发明的一些实施例，对夜光遥感影像进行可视化增强处理包括：对夜光遥感影像使用对数变换进行拉伸处理；对夜光遥感影像进行0~255可视化范围的拉伸处理。According to some embodiments of the present invention, performing visualization enhancement processing on the luminous remote sensing image includes: stretching the luminous remote sensing image using logarithmic transformation; and performing stretching processing on the luminous remote sensing image with a visualization range of 0-255.
根据本发明的一些实施例，对于原始的16bit夜光遥感影像，为了增加亮度信息的辨识度，采用对数函数进行拉伸处理，如下式所示：According to some embodiments of the present invention, for the original 16-bit luminous remote sensing image, in order to increase the recognition degree of brightness information, a logarithmic function is used to perform stretching processing, as shown in the following formula:
其中， p为夜光遥感影像的原始图像灰度值，L为对夜光遥感影像使用对数变换进行拉伸处理后的图像灰度值。lg(*)是取对数运算，max(*)是取最大值运算。Among them, p is the original image gray value of the luminous remote sensing image, and L is the image gray value after stretching the luminous remote sensing image using logarithmic transformation. lg(*) is the logarithm operation, and max(*) is the maximum value operation.
紧接着，对数变换后影像图L，进一步拉伸至0~255的可视化范围，即得到8bit夜光遥感数据，如下式所示：Next, after logarithmic transformation, the image map L is further stretched to the visualization range of 0~255, that is, 8bit luminous remote sensing data is obtained, as shown in the following formula:
其中， L为夜光遥感影像经过对数变换进行拉伸处理后的图像灰度值，V为对夜光遥感影像进行0~255可视化范围的拉伸处理后的图像灰度值。max(*)是取最大值运算，min(*)是取最小值运算。Among them, L is the gray value of the image after stretching the luminous remote sensing image through logarithmic transformation, and V is the gray value of the image after stretching the luminous remote sensing image with a visual range of 0 to 255. max(*) is the maximum value operation, and min(*) is the minimum value operation.
根据本发明的一些实施例，对夜光遥感影像进行可视化增强处理还可以使用基于对数拉伸、直方图拉伸、指数拉伸等方法。According to some embodiments of the present invention, methods based on logarithmic stretching, histogram stretching, exponential stretching, etc. may also be used to perform visualization enhancement processing on luminous remote sensing images.
S104，基于夜光遥感影像在原始夜光遥感底图上确定控制点，其中，控制点用于表征夜光遥感影像在原始夜光遥感底图上的对应位置。S104 , determining control points on the original night light remote sensing base map based on the night light remote sensing image, wherein the control points are used to represent the corresponding positions of the night light remote sensing image on the original night light remote sensing base map.
参见图2，根据本发明的一些实施例，本发明基于夜光遥感影像在原始夜光遥感底图上确定控制点包括：Referring to FIG. 2, according to some embodiments of the present invention, the present invention determines the control point on the original night light remote sensing base map based on the night light remote sensing image, including:
S1041，将夜光遥感影像划分为多个区块，在多个区块中选取对比度大于第一阈值的区块，作为特征匹配块。S1041: Divide the luminous remote sensing image into multiple blocks, and select blocks with a contrast greater than a first threshold from the multiple blocks as feature matching blocks.
夜光遥感影像仅获取夜间的高亮灯光和火光等信息，整景影像中大部分区域无有用信息。因此，为了提升后续模板匹配的稳定性和处理效率，首先将夜光遥感数据划分为离散的块区域，然后逐个计算分块区域的对比度，如下式所示：Night light remote sensing images only obtain information such as bright lights and firelights at night, and most areas in the whole scene image have no useful information. Therefore, in order to improve the stability and processing efficiency of subsequent template matching, the night light remote sensing data is first divided into discrete block regions, and then the contrast of the block regions is calculated one by one, as shown in the following formula:
其中，δ( i , j )=| i – j |，为相邻像素间的灰度差； P δ ( i , j )为向量像素间的灰度值为δ的像素分布概率。Among them, δ ( i , j ) = | i – j |, is the gray level difference between adjacent pixels; P δ ( i , j ) is the pixel distribution probability with the gray value of δ between vector pixels.
根据本发明的一些实施例，使用的像素相邻规则为四近邻或八近邻。According to some embodiments of the invention, the pixel neighbor rule used is four neighbors or eight neighbors.
分块对比度越高，说明影像包含的可识别信息越丰富，越适用于进行模板匹配处理。因此，本发明的一些实施例将对比度大于第一阈值的区块作为后续模板匹配的特征匹配块。The higher the block contrast, the richer the identifiable information contained in the image, and the more suitable for template matching processing. Therefore, in some embodiments of the present invention, a block with a contrast greater than the first threshold is used as a feature matching block for subsequent template matching.
根据本发明的一些实施例，在特征匹配块提取中，可以采用对比度方式计算特征匹配块的可匹配度，也可以采用例如灰度标准差、纹理丰富程度等方法进行度量。According to some embodiments of the present invention, in the feature matching block extraction, the matching degree of the feature matching block can be calculated by contrast, and can also be measured by methods such as gray standard deviation and texture richness.
S1042，将特征匹配块与原始夜光遥感底图进行模板匹配，确定特征匹配块相对于原始夜光遥感底图的第一平移量。S1042: Perform template matching between the feature matching block and the original night light remote sensing base map, and determine a first translation amount of the feature matching block relative to the original night light remote sensing base map.
根据本发明的一些实施例，确定特征匹配块相对于原始夜光遥感底图的第一平移量包括如下步骤：计算特征匹配块与原始夜光遥感底图在各平移量下的图像相似度系数；将图像相似度系数最大时对应的平移量确定为第一平移量；其中，计算特征匹配块与原始夜光遥感底图在各平移量下的图像相似度系数采用如下公式：According to some embodiments of the present invention, determining the first translation amount of the feature matching block relative to the original night light remote sensing base map includes the following steps: calculating the image similarity coefficient between the feature matching block and the original night light remote sensing base map under each translation amount; The corresponding translation amount when the image similarity coefficient is the largest is determined as the first translation amount; wherein, the following formula is used to calculate the image similarity coefficient between the feature matching block and the original night light remote sensing base map under each translation amount:
S coef (u,v)为图像相似度系数，(x,y)为原始夜光遥感底图位置坐标，(u,v)为特征匹配块相对于原始夜光遥感底图在(x,y)处的平移量，C(x-u,y-v)为特征匹配块在(x-u,y-v)处的图像灰度值，D(x,y)为原始夜光遥感底图在(x,y)处的图像灰度值，μ 1为特征匹配块的灰度均值，μ 2为原始夜光遥感底图的灰度均值，计算公式为：S coef ( u , v ) is the image similarity coefficient, ( x , y ) is the position coordinates of the original night light remote sensing base map, ( u , v ) is the feature matching block relative to the original night light remote sensing base map at ( x , y ) , C( x - u , y - v ) is the image gray value of the feature matching block at ( x - u , y - v ), D( x , y ) is the original night light remote sensing base map at ( x , y ), μ 1 is the gray mean value of the feature matching block, μ 2 is the gray mean value of the original luminous remote sensing base map, and the calculation formula is:
D(x-u,y-v)为原始夜光遥感底图在(x-u,y-v)处的图像灰度值，N为原始夜光遥感底图的像素数量。D( x - u , y - v ) is the image gray value of the original night light remote sensing base map at ( x - u , y - v ), and N is the number of pixels of the original night light remote sensing base map.
S1043，选取第一平移量在第一范围内的原始夜光遥感底图的图像点作为夜光遥感影像的控制点。S1043 , selecting the image points of the original luminous remote sensing base map with the first translation amount within the first range as the control points of the luminous remote sensing image.
控制点用于表征夜光遥感影像在原始夜光遥感底图上的对应位置。当将特征匹配块与原始夜光遥感底图进行模板匹配时，可以确定特征匹配块不同点数据在原始夜光遥感底图上的对应位置数据。由不同的数据进行模板匹配时，其确定的夜光遥感影像在原始夜光遥感底图上的对应位置数据会有不同。Control points are used to represent the corresponding positions of luminous remote sensing images on the original luminous remote sensing base map. When template matching is performed between the feature matching block and the original night light remote sensing base map, the corresponding position data of different point data of the feature matching block on the original night light remote sensing base map can be determined. When template matching is performed by different data, the corresponding position data of the determined luminous remote sensing image on the original luminous remote sensing base map will be different.
为了减小确定时的误差，本发明的一些实施例在计算出特征匹配块相对于原始夜光遥感底图的第一平移量时，会选取第一平移量在第一范围内的原始夜光遥感底图的图像点作为夜光遥感影像的控制点。也就是说，当计算出的第一平移量在第一范围内时，在原始夜光遥感底图上确定的夜光遥感影像的实际位置会更准确。In order to reduce the error during determination, some embodiments of the present invention will select the original luminous remote sensing base with the first translation within the first range when calculating the first translation amount of the feature matching block relative to the original luminous remote sensing base map. The image points of the map are used as the control points of the luminous remote sensing image. That is to say, when the calculated first translation amount is within the first range, the actual position of the luminous remote sensing image determined on the original luminous remote sensing base map will be more accurate.
本发明的一些实施例在得到特征匹配块与原始夜光遥感底图模板匹配的匹配结果后，会根据多个匹配结果采用RANSAC(RAndom SAmple Consensus,随机采样一致)算法来获得最佳的匹配位置，以减小确定位置的误差。In some embodiments of the present invention, after obtaining the matching results between the feature matching block and the original night light remote sensing base map template, the RANSAC (RANdom SAmple Consensus, random sampling consistency) algorithm is used to obtain the best matching position according to multiple matching results, in order to reduce the error in determining the position.
具体的步骤为：将第一平移量在第一范围内的原始夜光遥感底图的图像点标记为第一类点；采用RANSAC算法判断第一类点是否符合整体一致性；将符合整体一致性的第一类点作为夜光遥感影像的控制点。The specific steps are as follows: mark the image points of the original night light remote sensing base map with the first translation amount within the first range as the first type of points; use the RANSAC algorithm to judge whether the first type of points conform to the overall consistency; The first type of point is used as the control point of luminous remote sensing image.
S105，基于控制点，利用仿射变换模型，对夜光遥感影像进行校正。S105, based on the control points, using an affine transformation model to correct the luminous remote sensing image.
在上述步骤中选定控制点后，就可以基于控制点，利用仿射变换模型，对夜光遥感影像进行校正。After the control points are selected in the above steps, the affine transformation model can be used to correct the luminous remote sensing image based on the control points.
更具体地，本发明的一些实施例采用下述仿射变换模型方程，对夜光遥感影像进行校正：More specifically, some embodiments of the present invention use the following affine transformation model equations to correct night light remote sensing images:
其中，（X，Y）为控制点地面坐标，（x，y）为夜光遥感影像图上对应的坐标，a 0、a 1、a 2、a 3、b 0、b 1、b 2、b 3为仿射变换模型参数。Among them, ( X , Y ) are the ground coordinates of the control point, ( x , y ) are the corresponding coordinates on the night light remote sensing image map, a 0 , a 1 , a 2 , a 3 , b 0 , b 1 , b 2 , b 3 is the affine transformation model parameter.
当获取多个控制点数据后，例如获取控制点数量不少于4个，此时可以利用最小二乘法求出各仿射变换模型参数。After acquiring multiple control point data, for example, the number of acquired control points is not less than 4, the parameters of each affine transformation model can be obtained by using the least squares method.
然后进行坐标变换、重采样就可以利用上述仿射变换模型方程实现夜光遥感影像校正，获取精确的定位数据。Then, by performing coordinate transformation and resampling, the above-mentioned affine transformation model equation can be used to realize the correction of luminous remote sensing images and obtain accurate positioning data.
由于没有准确的夜光遥感底图数据，本发明的一些实施例为了提升夜光遥感底图的校正准确度，并保证夜光遥感底图的不断迭代更新，可以利用已校正的精确定位的夜光遥感影像数据更新夜光遥感底图。Since there is no accurate night light remote sensing base map data, some embodiments of the present invention can use the corrected and precisely positioned night light remote sensing image data in order to improve the correction accuracy of the night light remote sensing base map and ensure the continuous iterative update of the night light remote sensing base map. Update night light remote sensing basemap.
根据本发明的一些实施例，利用已校正的夜光遥感影像更新夜光遥感底图包括：基于已校正的夜光遥感影像的图像亮度，调整夜光遥感底图的像素灰度值，其中，对于已校正的夜光遥感影像中无光亮反应的区域，在夜光遥感底图中对应位置按第一比例减小像素灰度值，对于已校正的夜光遥感影像中有光亮反应的区域，在夜光遥感底图中对应位置按第二比例增大像素灰度值，第二比例基于已校正的夜光遥感影像的图像像素灰度值总和调整。According to some embodiments of the present invention, using the corrected luminous remote sensing image to update the luminous remote sensing basemap includes: adjusting the pixel gray value of the luminous remote sensing basemap based on the image brightness of the corrected luminous remote sensing image, wherein for the corrected luminous remote sensing image For the area without light response in the night light remote sensing image, the pixel gray value is reduced by the first proportion in the corresponding position in the night light remote sensing base map. The position increases the pixel gray value by a second ratio, and the second ratio is adjusted based on the sum of the image pixel gray value of the corrected luminous remote sensing image.
根据本发明的一些实施例，可以按照下述公式进行计算：According to some embodiments of the present invention, it can be calculated according to the following formula:
其中，D n为更新后的夜光遥感底图灰度值，C为已校正的夜光遥感影像灰度值，D为更新前的夜光遥感底图灰度值，max(*)为取最大值运算。Among them, D n is the gray value of the updated luminous remote sensing base image, C is the corrected gray value of the luminous remote sensing image, D is the gray value of the luminous remote sensing base image before the update, and max(*) is the operation of taking the maximum value. .
经过上述处理，可以根据不断累积的夜光遥感数据更新夜光遥感底图，将夜间高亮区域的灰度值增大，无亮度反应区域的灰度值减小，逐渐接近真实的夜间遥感获取结果。After the above processing, the night light remote sensing base map can be updated according to the accumulated night light remote sensing data, the gray value of the nighttime highlight area is increased, and the gray value of the non-brightness response area is decreased, gradually approaching the real nighttime remote sensing acquisition results.
如图3为本发明的一些实施例中实施夜光遥感影像校正方法的整体流程示意图。在这些实施例中，本发明实现夜光遥感数据的自动化精确定位，同时通过夜光遥感观测数据的累积获得更精确的夜光遥感底图数据，可以逐渐优化夜光遥感底图数据，有利于夜光遥感影像的高精度定位处理。FIG. 3 is a schematic diagram of an overall flow of implementing a method for calibrating a night light remote sensing image in some embodiments of the present invention. In these embodiments, the present invention realizes automatic and precise positioning of night light remote sensing data, and simultaneously obtains more accurate night light remote sensing base map data through the accumulation of night light remote sensing observation data, which can gradually optimize the night light remote sensing base map data, which is conducive to the improvement of night light remote sensing images. High-precision positioning processing.
以上所述的具体实施例，对本发明的目的、技术方案和有益效果进行了进一步详细说明，所应理解的是，以上所述仅为本发明的具体实施例而已，并不用于限制本发明，凡在本发明的精神和原则之内，所做的任何修改、等同替换、改进等，均应包含在本发明的保护范围之内。The specific embodiments described above further describe the purpose, technical solutions and beneficial effects of the present invention in further detail. It should be understood that the above descriptions are only specific embodiments of the present invention, and are not intended to limit the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.
Claims (10)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210364692.2A CN114638766B (en) | 2022-04-08 | 2022-04-08 | Method for correcting luminous remote sensing image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210364692.2A CN114638766B (en) | 2022-04-08 | 2022-04-08 | Method for correcting luminous remote sensing image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114638766A true CN114638766A (en) | 2022-06-17 |
| CN114638766B CN114638766B (en) | 2022-11-22 |
Family
ID=81951035
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210364692.2A Active CN114638766B (en) | 2022-04-08 | 2022-04-08 | Method for correcting luminous remote sensing image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114638766B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116630204A (en) * | 2023-07-19 | 2023-08-22 | 南京佳格耕耘科技有限公司 | Remote sensing image online analysis processing system |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105930867A (en) * | 2016-04-19 | 2016-09-07 | 中国科学院遥感与数字地球研究所 | Data mining method based on time series luminous remote sensing image |
| US20170039436A1 (en) * | 2015-08-03 | 2017-02-09 | Nokia Technologies Oy | Fusion of RGB Images and Lidar Data for Lane Classification |
| CN106920235A (en) * | 2017-02-28 | 2017-07-04 | 中国科学院电子学研究所 | Star-loaded optical remote sensing image automatic correction method based on the matching of vector base map |
| CN107341781A (en) * | 2017-07-20 | 2017-11-10 | 中国科学院电子学研究所 | Based on the SAR image correcting methods for improving the matching of phase equalization characteristic vector base map |
| CN109977344A (en) * | 2019-03-20 | 2019-07-05 | 武汉大学 | A kind of block adjustment method of spaceborne noctilucence remote sensing image |
| CN113610873A (en) * | 2021-10-09 | 2021-11-05 | 中国科学院地理科学与资源研究所 | Noctilucent remote sensing image data processing method and device |
| CN113723175A (en) * | 2021-07-14 | 2021-11-30 | 中国人民解放军战略支援部队信息工程大学 | Method for extracting urban built-up area of remote sensing image |
-
2022
- 2022-04-08 CN CN202210364692.2A patent/CN114638766B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170039436A1 (en) * | 2015-08-03 | 2017-02-09 | Nokia Technologies Oy | Fusion of RGB Images and Lidar Data for Lane Classification |
| CN105930867A (en) * | 2016-04-19 | 2016-09-07 | 中国科学院遥感与数字地球研究所 | Data mining method based on time series luminous remote sensing image |
| CN106920235A (en) * | 2017-02-28 | 2017-07-04 | 中国科学院电子学研究所 | Star-loaded optical remote sensing image automatic correction method based on the matching of vector base map |
| CN107341781A (en) * | 2017-07-20 | 2017-11-10 | 中国科学院电子学研究所 | Based on the SAR image correcting methods for improving the matching of phase equalization characteristic vector base map |
| CN109977344A (en) * | 2019-03-20 | 2019-07-05 | 武汉大学 | A kind of block adjustment method of spaceborne noctilucence remote sensing image |
| CN113723175A (en) * | 2021-07-14 | 2021-11-30 | 中国人民解放军战略支援部队信息工程大学 | Method for extracting urban built-up area of remote sensing image |
| CN113610873A (en) * | 2021-10-09 | 2021-11-05 | 中国科学院地理科学与资源研究所 | Noctilucent remote sensing image data processing method and device |
Non-Patent Citations (2)
| Title |
|---|
| JINHU BIAN 等: ""Inter-Calibration of Nighttime Light Data Between DMSP/OLS And NPP/VIIRS in the Economic Corridors of Belt And Road Initiative"", 《IEEE》 * |
| 李明峰 等: ""NPP/VIIRS多时相夜光遥感影像校正方法"", 《测绘通报》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116630204A (en) * | 2023-07-19 | 2023-08-22 | 南京佳格耕耘科技有限公司 | Remote sensing image online analysis processing system |
| CN116630204B (en) * | 2023-07-19 | 2023-09-26 | 南京佳格耕耘科技有限公司 | Remote sensing image online analysis processing system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114638766B (en) | 2022-11-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107563964B (en) | Rapid splicing method for large-area-array sub-meter-level night scene remote sensing images | |
| CN110246100B (en) | An image inpainting method and system based on angle-aware block matching | |
| CN106920235B (en) | Automatic correction method of spaceborne optical remote sensing image based on vector basemap matching | |
| CN111932504B (en) | Edge contour information-based sub-pixel positioning method and device | |
| CN109741446B (en) | Method for dynamically generating fine coast terrain by three-dimensional digital earth | |
| CN114998545A (en) | Three-dimensional modeling shadow recognition system based on deep learning | |
| CN110136177A (en) | A kind of method for registering images, equipment and storage medium | |
| CN113569647A (en) | AIS-based ship high-precision coordinate mapping method | |
| CN114638766B (en) | Method for correcting luminous remote sensing image | |
| CN109671038B (en) | A Relative Radiometric Correction Method Based on Pseudo-Invariant Feature Point Classification and Layering | |
| CN111220235A (en) | Water level monitoring method and device | |
| CN108629742B (en) | True ortho image shadow detection and compensation method, device and storage medium | |
| CN108364320A (en) | camera calibration method, terminal device and computer readable storage medium | |
| CN118918482B (en) | Natural resource measurement method and system based on remote sensing images | |
| CN114782824A (en) | Wetland boundary defining method and device based on interpretation mark and readable storage medium | |
| CN108917722A (en) | Vegetation coverage calculation method and device | |
| CN117451012B (en) | Unmanned aerial vehicle aerial photography measurement method and system | |
| CN111275698B (en) | Method for detecting visibility of road in foggy weather based on unimodal offset maximum entropy threshold segmentation | |
| CN113592744A (en) | Geometric precise correction method suitable for high-resolution remote sensing image | |
| CN113283326A (en) | Video SAR target intelligent detection method based on simulation target bright line characteristics | |
| CN110322454B (en) | High-resolution remote sensing image multi-scale segmentation optimization method based on spectrum difference maximization | |
| CN113781587B (en) | Remote sensing image color consistency processing method based on optimal path | |
| CN113034555B (en) | Feature fine matching method based on minimum spanning tree and application | |
| CN116403011A (en) | Method for establishing effective data sample group based on SAR-Optical image matching | |
| CN115330618A (en) | Local orthorectification method based on video satellite images and range vectors |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |