Aerial filter array multispectral images and their high precision registrations are important for guaranteeing subsequent image processing and application.In the process of image registration,the position accuracy of matching points is important in determining the accuracy of image registration.However,objects of different strips in the same band image are acquired at different moments,the image displacement between single-band images is large,and the difference in geometric errors of matching points between topographic undulating areas and flat areas in the image is obvious.Additionally,false matching points cannot be accurately eliminated by the global matrix.The difficulty of eliminating mismatched points in multispectral images of aerial filter arrays must be addressed because of the displacement of image points between spectral segments.Thus,a new method of double-threshold elimination based on matching point position difference surface fitting is proposed in this study. First,the intermediate band image of the filter array multispectral images was selected as the reference image,and the matching points in the reference image and the image to be registered were extracted by the subpixel-level SIFT algorithm.Second,the difference in the positions of the matching points of the two bands was calculated point by point at the matching points of the benchmark image,and the Delaunay triangulation network of matching points in the reference image was constructed.The position difference surface was smoothed,the position differences between the matching points of the reference image and the corresponding matching points of the image requiring registration were calculated point by point,and a certain tolerance range was shifted upward and downward to form a 3D position difference threshold space.Finally,accurate matching points were selected using the 3D threshold space of the position difference to complete the registration. The three-band composite image of the algorithm-registered image in this study presented clear features and well-defined details and met the requirements of subsequent data processing and application.The effectiveness of the proposed algorithm was illustrated by registering two datasets of filter array multispectral images,from which qualitative and quantitative perspectives were verified.Regarding false color,the composite image processed by the proposed algorithm did not show obvious pseudoedges,and the features were clear.However,pseudoedges were obvious in the comparison algorithm and difference image grayscale histograms.Among the experiments of the two datasets,the difference image histogram curve of the proposed algorithm presented the largest shift to the left.The image registered by the proposed algorithm had the smallest difference from the reference image and the best registration effect. Theoretical analysis and experimental results show that the dual-threshold pointing algorithm based on matching point difference fitting of curved surfaces can screen high-precision matching points in aerial filter array multispectral images and effectively improve the accuracy of image registration.Surface fitting to the position difference of matching points can help reveal the trend of image point displacement in each region.This scheme can also effectively eliminate false matching points around the correct matching points,especially since the image displace.

针对航空滤光片阵列多光谱图像条带重叠率低与条带边缘光谱混叠干扰导致条带有效区域裁剪不精准问题,提出航空滤光片阵列多光谱图像条带预处理算法.通过航空滤光片阵列多光谱图像行像素灰度均值,依次计算条带图像间的灰度极值点;条带图像中间区域行像素灰度均值的最小值与其行像素灰度变化曲线相交位置作为条带截止点,对灰度极值点与条带截止点对应的行方向坐标进行作差,得到条带图像边缘光谱混叠范围;根据各条带图像边缘最大光谱混叠宽度和相邻灰度极值点行方向坐标,依次计算各条带有效区域的顶点坐标,通过图像裁剪得到各单波段条带图像.理论分析与实验结果表明:该方法可以精确提取各条带图像有效区域,且能够最大程度保留重叠区域像素,裁剪后的各条带有效区域像素灰度变化一致,不存在光谱混叠现象.

Filter array multispectral cameras are influenced by imaging mechanism and process characteristics, spliced images have edge interference fringes and greyscale differences. Aiming at the problems of inconsistent greyscale of filter array multispectral camera, a new method of greyscale correction algorithm is proposed in this paper. First, the mechanism and adjustment principle of edge interference streaks and stripe greyscale difference are thoroughly analyzed; Second, the greyscale of interference area is adjusted by using greyscale of adjacent image non-interference area; Third, the greyscale of whole image is adjusted by using proportional relationship between the adjacent overlap area greyscale; Finally, sequence images are spliced to obtain single-band image with same greyscale. Theoretical analysis and experimental results show that this method can not only effectively solve the problem of inconsistent greyscale due to the influence of imaging mechanism and process characteristics, but also can maximally preserve spectral information characteristics in different wavelength bands.

针对倾斜遥感图像拼接中存在配准精度不高,重叠区域出现重影的问题,提出了利用三角网进行小面元配准及加权融合的拼接方法.首先,求解全局单应性矩阵进行图像预对齐,并利用转换矩阵提取图像重叠区域;其次,利用特征匹配对构建重叠区域Delaunay三角网并对相应三角网逐个进行仿射变换实现精确配准;最后,利用渐出渐入式融合消除图像重叠区域重影,提高了图像拼接目视效果.通过主观评价(CCIR500-1的主观评价标准)与客观评价[峰值信噪比(peak signal-to-noise ratio,PSNR)评价和结构相似性(structural similarity index,SSIM)评价]相结合的方式对图像配准及拼接质量进行评价.实验结果表明:该方法能够有效提高图像配准精度,消除图像拼接重叠区域的重影,相比于当前主要算法,该算法配准精度更高,图像拼接目视效果更好,能有效地应用到倾斜图像拼接领域.

针对航空滤光片阵列多光谱图像中易出现条带灰度不一致问题，提出条带灰度调整算法。利用状态信息对图像进行投影变换，利用SIFT算法提取图像匹配点，计算匹配点坐标差均值作为图像间的平移关系；基于波段图像的重叠区域计算灰度平均值，利用各波段灰度均值比的平均值作为图像灰度调整系数；以中间图像灰度为基准，对图像进行灰度调整，再通过图像裁剪与拼接得到各单波段灰度一致图像。理论分析与实验结果表明：该方法不仅可以有效解决滤光片阵列多光谱图像灰度不一致的问题，而且能够最大限度保持地物光谱信息特征。

Aiming at the problem of low image stitching accuracy due to the difficulty of extracting strip features from multispectral images of filter array, a strip stitching algorithm for multispectral images of filter arrays is proposed. Firstly, the effective range of each band of the multispectral image is determined, and the homography matrix is calculated using the geographic coordinate information of the image vertices and the vertex coordinates to project the image. Secondly, Scale-invariant feature transform (SIFT) algorithm was used to extract matching points of projected images, and the mean value of the coordinate difference of matching points was calculated as the translation relation between images. Finally, the projection transformation is performed on the single band bands in turn, and the projected images are stitched with inter-image translation to obtain a large area single band image. Theoretical analysis and experimental results show that this method can effectively improve the the stitching accuracy of multispectral images of filter arrays and has a high image stitching speed.

提出了一种基于新的空间坐标系统的拼接方法.定义了机北坐标系统和机平坐标系,推导了机北坐标系与机平坐标系、载机坐标系间转换关系,改进了几何校正算法;研究了图像像底点和地底点的计算方法,并利用空间地理坐标信息进行粗拼接;利用特征匹配进行坐标微调,提出了三步进图像坐标微调策略,实现了航空三步进分幅图像精拼接.实验结果表明:该方法在处理多条带图像时拼接效果较好,提出方法对三步进分幅图像的拼接精度较现有方法更高,具有较强的实用性.

色彩匹配函数使得多光谱数据合成彩色图像简单易行,但是直接使用色彩匹配函数加权得到的多光谱彩色图像容易偏色,且无法避免低光照度和阴影对图像质量的影响;针对这一问题,本文结合色彩恒常理论,使用双sigmoid函数作为权重优化了Frankle-McCann Retinex算法在像素亮度计算的重置策略,达到对基于颜色匹配函数的真彩色复原技术改进,实现地物的彩色复原和增强;首先,对各波段图像建立亮度比较路径,通过对路径像素计算和迭代,估计目标在各波段的亮度值,其次利用色彩三刺激值原理与多光谱相机系统参数建立彩色图像三刺激分量生成模型,最后通过色彩空间转换得到目标真实色彩;利用Cave实验室数据和航空多光谱数据进行了验证和评价,结果表明本文提出的方法能够充分利用多光谱相机各波段图像信息,能够对图像阴影部分的目标色彩增强,色彩复原效果相对传统方法有明显提升,对于航空多光谱图像目视解译工作具有重要意义.

A method of airport target detection and analysis of Synthetic Aperture Radar image is suggested in this article based on the facts. The airport detection processing steps are presented and the target shape analysis method is suggested. The airport target is automatically detected and the parameters of target are calculated. Experimental results show that the method is efficient

美军重视院校教育对联合军官的培养,本文首先分析了美军院校联合职业军事教育培养体系的教学层次分明和教学目标明确的特点,之后重点从课堂教学要素的角度论述了美军在教学内容、教员队伍、学员组成和教学方法方面的做法,以前对我军联合人才培养提供一定借鉴.

实践教学是培养高素质创新人才的重要环节.结合任职教育人才培养实践教学体系建设的实际情况,阐述了实践教学在任职教育中的重要作用,分析了目前任职教育实践教学中存在的问题,提出了任职教育实践教学体系建设的对策思考.

To solve the problem that constructing local feature descriptors of SURF takes too much time, this paper proposed a novel method based on binary BRISK feature descriptor to improve remote sensing image matching speed. Firstly fast DoH (Determinant of Hessian) operator was applied to detect the feature points and determine the main direction. Then we used BRISK descriptor to describe the feature points, at the matching stage, we used two direction matching by Hamming distance and RANSAC algorithm to match again. Finally, the least square algorithm was applied to obtain accurate registration and complete image registration. Experimental results show that the algorithm can not only improve the matching speed significantly, but also the matching performance is beyond the original SURF algorithm, which can be used to complete the remote sensing image registration.

In order to increase the accuracy of locating the object by UAV, a robust algorithm based on image-reconnaissance is proposed. First, to record real-time objects reconnoitered by the photovoltaic system on the image. Then, to combines the target's coordinates on the image with UAV flight parameters. Finally, through coordinate transformation, geometric calculation and other processes, it is able to locate the target's geodetic coordinates. Monte Carlo simulation is introduced to the simulation experiments, to prove that this targeting method can meet the requirements of practical application. Experimental results show that the proposed algorithm has a good real-time performance, accuracy and reliability.

In view of the problem that current methods cannot reach a good balance between capability of discrimination, utility and computational complexity, the authors have proposed in this paper an algorithm based on hierarchical feature description. Firstly, simple shape or geometrical features are extracted to get rid of large numbers of false-alarm targets based on weighted voting. Secondly, complex discrimination features are selected to form the optimal feature set by feature separation. And then the feature set is used to support vector machine to get the real ship target. Experimental results show that the proposed algorithm in this paper, which extracts hierarchical features to certain regions identified, can effectively eliminate false alarms, reduce the amount of computation, and improve accuracy and efficiency of discrimination, and can also reduce the influence of external factors, remove false alarm and reserve the targets effectively, with time spending being only 1/3 of the common method.

UAVs can locate the precise location information of targets through a variety of methods in order to implement battlefield command or a military strike. The various sorting and classification principles of UAVs reconnaissance targeting technology are described, and its scope of application in the military field is parsed. UAV mission is taken as an example, in which a detailed calculation of the amount of space UAV reconnaissance targeting technique is performed by using the actual flight path planning of UAV reconnaissance, which proves targeting technology has an important military value.

In order to improve the targeting UAV speed and precision, a reliable image-based reconnaissance drones targeting algorithm is presented. Firstly, the target image information is gained by the photoelectric system of UAV, and then the target’s coordinates in the image and UAV flight parameters are combined to calculate the geodetic coordinates of targets with the use of the homogeneous coordinate transformation method. During the entire localization process, the actual errors are taken into consideration, and the Monte Carlo analysis method is applied to analyze the error according to UAV flight recorder data. Experimental results show that this algorithm can quickly and accurately locate the object, and the positioning accuracy is 12.683195 m, which can illustrate the algorithm in real-time, accuracy, reliability and feasibility.

针对遥感图像中机场跑道区域比较狭长的特点,提出一种新的检测遥感图像中机场跑道的方法.跑道作为机场最显著的标志是由于它具有明显的直线特征和边缘特征,本文提出基于Hough变换的改进算法,采用直线支持区间的思想,将相位角的范围缩小到一个小的区域进行检测,大大减少了计算量.然后采用直线连接的方法将满足一定条件的、存在断点的线段连接,最后根据跑道特点,去除不相关的平行线,最终检测出机场跑道,实现了快速精确.

In order to solve the problem that majority of ship detection algorithms show the low accuracy and slow rate, a fine rapid ship detection algorithm based on salient feature guidance is proposed. First, the candidate target area is located through visual saliency model based on the local and global integration, and candidate target slices are gained through the region extraction. Then, the improved means clustering method is used to divide the target slice into super pixels. Finally, ship target regions are gained through filtering super pixels belonging to the target to achieve a fine segmentation after the integration of a significant figure and super-pixel segmentation. Experimental results show that the algorithm proposed in this paper can quickly and accurately locate the target ship, accurately depict the object contour, and is more conducive to subsequent follow-up work.

This paper discusses techniques for image alignment software development based on OpenCV. Image alignment technology is a key point of full view imaging and remote sensing technology, and the price of conventional image alignment software is expensive. This paper developed a new software environment based on open source computer image library OpenCV, it can rapidly optimization upon the different image alignment demands. Using the image class can setup the interface of static images or dynamic images, corner points detecting and sifting feature points, then calculating the alignment factors and results of image alignment. Furthermore, using stitching class can calculate the image stitching results to check the image alignment results. Experiments show that this software can calculate complex image alignment results, checking the alignment goodness and with less time cost.

Normalized cross correlation algorithm for image matching is a frequently-used image matching algorithm which based on gray correlation, but the algorithm has higher computational cost and lower speed. The paper is based on studying the correlation coefficient formula of the normalized cross correlation algorithm, and the purpose of the paper is to reduce the computational cost and make the matching speed faster.