RESEARCH AND DESIGN OF INSPECTION CLOUD PLATFORM FRAMEWORK FOR SURVEYING AND MAPPING PRODUCTS

With the continuous improvement of modern surveying and mapping technology and with the plentiful of achievements, traditional quality inspection software for single machine, single task and single data type, difficult to massive multi-source isomerization achievements, difficult to meet the requirement of rapid, accurate and efficient quality inspection. With the development of IT technology such as cloud computing, big data and artificial intelligence, the quality inspection software needs to combine cloud computing technology with quality inspection business, refactoring software framework. Facing to the storage and spatial query requirement of inspection for surveying and mapping products, the paper researches and designs the spatial data distributed storage and the spatial data distributed index in cloud platform. The Management of inspection rule is the core in cloud platform. Inspection rule is the minimum operating independent unit, which becomes inspection item by parameterization, the paper builds full run-time operating mechanism in cloud platform for inspection rule. Finally, Combining the inspection requirement for surveying and mapping products and business, the paper researches and design the cloud framework for surveying and mapping products.


Geographic Information Quality Inspection
Hadoop, Spark is more convenient and efficient in large-scale data processing, and it has widespread recognition and support. But, mature solution in the field of surveying and mapping geographic information is not exist, and the same situation was happened in the field of surveying and mapping geographic information quality inspection. Based on the analysis of the operation problems of current surveying and mapping geographic information quality inspection software, this paper combines the quality inspection business requirements with cloud computing, researches and designs a cloud platform for surveying and mapping geographic information quality inspection based on Hadoop framework, Spark computing engine and distributed database MongoDB.

Spatial data storage and management
Storage and management of geospatial information data has experienced File-RDBMS, RDBMS, ORDB, OODB. (Chen GuoPing, 2013) In the era of huge data with increasing demand for geographic information data, the above four methods can't solve the problem of storage and management of massive spatial data well. Distributed database and distributed file system came into being. Massive spatial data was decomposed into smaller data blocks and distributed to each node for storage, which greatly shortened the time of data storage process.

Spatial data indexing
Spatial data indexing arises at the historic moment in spatial database technology. Its main goal is to speed up the system's retrieval of spatial data. It can improve the retrieval speed by reducing irrelevant data or clustering related data. Dozens of spatial

MongoDB
The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume XLII-3/W10, 2020 International Conference on Geomatics in the Big Data Era (ICGBD), 15-17 November 2019, Guilin, Guangxi, China

MongoDB is a distributed file system-based
NoSQL open source database project that uses loose storage similar to JSON to store more complex data types, with indexing, sharding, load balancing, aggregation, and more (Zhou Yao, 2018

Construction technology of quality inspection rule library based on spark
The quality inspection rule is an abstraction of possible data defects. It has universal applicability, is the smallest unit of the quality inspection model, and is the most dynamic quality inspection element. It is the entity that carries the quality inspection function.
In the quality inspection process, the quality inspection function of the software is the embodiment of the quality inspection rules. The quality inspection rules are not static. They may increase due to the increase in the type of data, or may be adjusted due to the adjustment of data standards.
The quality inspection rule is instantiated by parameter activation to form an inspection item. The inspection item is a rule that is used to check whether a data has a specific error, and is the result of the rule instantiation. A complex inspection scheme can be completed by the logical combination of inspection items.  The action rule triggers Spark to submit the job, mainly including reduce, collect, count, etc.

Framework design
The inspection cloud platform framework for surveying and mapping products is divided into 3 layers.  The data layer provides data storage and access.

Data
The data contains four types, including spatial data,