基于多特征的高分遥感图像分割算法研究

针对传统的图像分割算法不能完全适用具有多种特征〖BF〗（〖BFQ〗光谱特征、纹理特征和几何特征〖BF〗）〖BFQ〗的高分辨率遥感图像的问题,提出了一种基于多特征的遥感图像分割算法。算法基于改进的均值漂移滤波和自动标记分水岭分割方法来实现最终分割。首先利用自动标记分水岭分割方法对遥感图像进行分割,进而采用仿射不变矩形状特征算子提取图像几何特征;其次对图像进行主成分分析,计算第一主成分灰度共生矩阵,分析矩阵特性得出纹理特征;然后结合光谱特征通过改进的均值漂移方法得到多特征滤波结果;最后利用分水岭分割方法实现高分辨率遥感图像分割。为了表明算法的分割效果,利用基于多光谱信息熵方法对算法和单一的分水岭分割方法进行非监督评价。研究结果表明,算法可较好地改善遥感图像的过分割问题,是一种适合高空间分辨率多光谱遥感图像的分割算法。     

基于图像的经纬仪测量目标快速瞄准技术

电子经纬仪是一种高精度的测角仪器,由多台高精度电子经纬仪构成的测量系统可以实现三维坐标测量,用于对大型工业产品的几何检测。但经纬仪在测量过程中仍然需要人工瞄准目标,存在速度慢、效率低、人为因素影响大等不足。实现经纬仪自动照准目标功能将是一个新的发展方向。提出一种基于图像的自动瞄准技术,通过在经纬仪上安装一台相机获取待测目标点的图像,利用图形目标识别技术获取目标点图像坐标,依据相机与经纬仪的空间关系可以获得各目标点相对于经纬仪的初始角度,从而实现目标点的快速概略定位,理论推导及实验证明本方法能够准确有效的获得目标点的概略位置。     

微小刀具同轴全息对刀方法研究

针对现有对刀方法存在的各种不足,提出了一种微小刀具的同轴全息对刀方法。在对全息成像原理和全息图像关键问题进行研究与分析的基础上,搭建了一套微细铣刀的全息对刀检测装置,并利用标准刀具在五轴加工中心进行了微细铣刀的对刀测量试验。试验结果表明:该对刀装置能够高效地进行微细铣刀的对刀检测,系统相对检测精度达5.1μm。在相同检测条件下,与高精度机外预调仪的测量结果相比,接触式对刀仪的相对误差为0.033%,而全息对刀样机的相对误差为0.007%,后者更能有效地实现微小刀具的对刀检测。该结果验证了微小刀具的同轴全息对刀方法的可行性,即全息测量可用于微细铣刀的高精度对刀。     

地形纹理分割技术在雷达图像仿真中的应用

文章阐述了将图像分割技术应用于雷达地形纹理分割中,用基于 FCM 图像分割方法对可见光纹理进行区域自动识别,形成互不相交的纹理区域,并借助建立的雷达材质库对区域材质进行指定,有效解决了海量地形纹理数据处理的难题,并提高了处理的准确性.     

基于纹理的活动轮廓三维图像分割研究

对具高噪声和低对比度三维图像的识别和分割算法进行了研究。基于活动轮廓模型,用Gabor变换提取图像的纹理特征,根据统计学信息假设,通过偏微分方程水平集和窄带方法求解,获得较基本活动轮廓的算法分割更光滑精确的物体轮廓。实验结果表明:改进模型算法的效率和准确度达到预期效果。     

星载微波辐射计通道分辨率匹配技术及其在FY-3卫星MWRI中的应用

基于仿真模型对圆锥扫描星载微波成像仪不同频率和极化二维天线方向图地面足迹的模拟,对FY-3卫星微波成像仪(MWRI)原始观测轨道亮温数据,用Backus-Gilbert(BG)分辨率匹配算法分别对分辨率降低(高频通道匹配到低频通道)和分辨率增强(低分辨率匹配到高分辨率)进行了实验分析。结果表明:分辨率降低匹配处理能较好地模拟实际仪器观测,且不另引入噪声,可在MWRI通道分辨率匹配中实际应用;分辨率增强匹配可明显提高低频图像分辨率,但会一定程度引入噪声,实际应用中应考虑在分辨率增强与噪声间进行折中处理,在增强分辨率的同时,使引入的噪声在可接受范围内。     

控制点/控制点对最优能量解析的图像配准

 针对遥感图像配准中匹配同名点少、配准精度低等问题,提出了一种基于控制点/控制点对最优能量解析的图像配准方法。在控制点提取中,通过多尺度曲波变换,实现不同尺度下图像质量的灰关联分析和噪声抑制,并根据灰关联能量值,自动调整尺度不变特征变换配准参数,提高匹配同名点数量;为保证控制点选择的均匀性和准确性,创建了控制点对能量模型,采用马尔科夫蒙特卡洛优化算法实现分布、匹配能量最优的控制点筛选,提高了图像配准精度。实验结果表明,本方法对遥感图像匹配同名点少、控制点分布不均匀、图像配准精度低等问题有很大改善,具有较高的应用价值。     

探空火箭图像压缩算法研究

结合探空火箭图像系统需求, 提出了一种基于小波变换编码, 即基于MSPIHT的自适应差分算术编码(MSPIHT Adaptive Differential Arithmetic Coding for image, MADAC)的图像压缩算法. 该算法对经过量化编码之后的数据 进行熵编码. 实验结果表明, MADAC算法具备探空火箭图像系统所要求的分辨 率、帧速率及压缩率可调的功能, 且易于硬件实现.      

Aerial-BiSeNet: A real-time semantic segmentation network for high resolution aerial imagery

The aircraft system has recently gained its reputation as a reliable and efficient tool for sensing and parsing aerial scenes. However, accurate and fast semantic segmentation of high-resolution aerial images for remote sensing applications is still facing three challenges: the requirements for limited processing resources and low-latency operations based on aerial platforms, the balance between high accuracy and real-time efficiency for model performance, and the confusing objects with large intra-class variations and small inter-class differences in high-resolution aerial images. To address these issues, a lightweight and dual-path deep convolutional architecture, namely Aerial Bilateral Segmentation Network (Aerial-BiSeNet), is proposed to perform real-time segmentation on high-resolution aerial images with favorable accuracy. Specifically, inspired by the receptive field concept in human visual systems, Receptive Field Module (RFM) is proposed to encode rich multi-scale contextual information. Based on channel attention mechanism, two novel modules, called Feature Attention Module (FAM) and Channel Attention based Feature Fusion Module (CAFFM) respectively, are proposed to refine and combine features effectively to boost the model performance. Aerial-BiSeNet is evaluated on the Potsdam and Vaihingen datasets, where leading performance is reported compared with other state-of-the-art models, in terms of both accuracy and efficiency.     

Effect of scavenge port angles on flow distribution and performance of swirl-loop scavenging in 2-stroke aircraft diesel engine

Swirl-Loop Scavenging (SLS) improves the performance of 2-stroke aircraft diesel engine because the involved swirl may not only benefit the scavenging process, but also facilitate the fuel atomization and combustion. The arrangement of scavenge port angles greatly influences in-cylinder flow distribution and swirl intensity, as well as the performance of the SLS engine. However, the mechanism of the effect and visualization experiment are rarely mentioned in the literature. To further investigate the SLS, Particle Image Velocimetry (PIV) experiment and Computational Fluid Dynamics (CFD) simulation are adopted to obtain its swirl distribution characteristics, and the effect of port angles on scavenging performance is discussed based on engine fired cycle simulation. The results illustrate that Reynolds Stress Turbulence model is accurate enough for in-cylinder flow simulation. Tangential and axial velocity distribution of the flow, as well as the scavenging performance, are mainly determined by geometric scavenge port angles α_geom and β_geom. For reinforcement of scavenging on cross-sections and meridian planes, α_geom value of 27° and β_geom value of 60° are preferred, under which the scavenging efficiency reaches as high as 73.7%. Excessive swirl intensity has a negative effect on SLS performance, which should be controlled to a proper extent.