Remote sensing images are the true representation of various natural and human elements in the surface scene. Due to their good adaptability and wide detection range, they are widely used in the military field, and play an important role in fire reconnaissance, intrusion warning, flight guidance, landform mapping, etc. important role. For a remote sensing image, how to quickly and efficiently complete its feature evaluation has gradually become a hot research issue in recent years. This paper builds a set of visual parallel processing remote sensing image evaluation system. The system can support three analysis modes: pure terrain, pure scene, and scene plus terrain. The grid window is the basic unit, and multiple CPUs are used to perform parallel feature extraction and analysis on remote sensing images, and mirror and display the evaluated results in the $3\mathrm{D}$ scene. Remote sensing image data. Compared to traditional methods, software significantly improves computational efficiency and result availability.

Space launch failure is an uncommon but extremely important real-world problem. This paper proposes a Hierarchical Timed Colored Petri nets (HTCPN) simulation and validation method for the emergency response plan in response to the fact that the emergency response process of space launch failure is difficult to be fully practiced in real launch scenarios. Firstly, HTCPN is introduced as a kind of advanced petri net which is convenient in modeling and analyzing the actual complex system. Then the HTCPN model is constructed for the key process links of space emergency launch, and finally the model is dynamically operated and its performance is evaluated. After analysis, it can be seen that the use of HTCPN to model the space launch emergency response plan can summarize the disposal process comprehensively, describe it clearly, and analyze the results with credibility.

针对海量装备试验数据结构复杂、分散存储及检索访问等问题,开展装备试验数据融合存储技术研究.提出了数据源插件技术、分布式采集处理技术、海量数据文件索引技术,实现多源异构试验数据的采集整理、校核验证、入库和多维信息查询,以及图像、视频、文档类数据文件的检入、检出等功能,并保证系统稳定、高效运行,确保数据应用业务可扩展、可发展.该装备试验数据融合存储技术已应用到某装备系统中,为系统试验数据的稳定存储和应用提供了基础支撑,也为多源异构试验数据的融合存储和管理提供了一种解决方案.

随着航天各专业技术的快速发展,专家技术评审已成为型号质量工作中的一个重要环节,通过专家团队集中把关,可有效发现产品设计缺陷.但是,目前的专家技术评审工作依然采用召开评审会的方式,依靠人工的形式传递文档、组织会议、通知专家、汇总结论,整个过程中消耗了大量的人力、物力和时间.随着型号数量的增长,技术评审的数量和要求急剧增多,传统的评审模式已不能适应当前的需求,导致技术评审的效率和质量不断下降.不仅如此,一些评审活动流于形式,有效性不足.

Efficient and rapid recognition of aircrafts is of great significance in the fields of air traffic control, airport monitoring and military surveillance. In this paper, a method of fusion moment invariant aircraft recognition (FMIAR) on multi-source information is proposed, which can deal with the multisensor images in real-time and eliminate the influences of image translation, scaling, rotation and background noise. Furthermore, unreliable sensors can be cast off tactically to identify the aircraft type. Experiments show that FMIAR satisfies requirements of effective real-time online identification.

The space flight software is one of the most important parts of the spacecraft control system,and the 1553B bus is more and more extensively used in the spacecraft.Thus,the simulation for the 1553B bus environment is indispensable and increasingly critical.However,due to the independence of the different types of space flight software,each type needs to design the specific simulation test platform.A general simulation test platform(GSTP) is presented and the characteristics and the test function module of 1553B bus based on this platform are studied,then the detailed design of the requirement and structure is separately implemented on the driver layer,logic layer and user layer.This technique can be commonly used in any kind of space flight software that supports the 1553B bus,which significantly reduces the software test costs and effectively shortens the test cycle.