LAND COVER CHANGE AUTOMATIC DETECTION BASED ON 
PARCEL-KNOWLEDGE INTEGRATED RS WITH GIS 
YANG Guijun 
Dept. of Surveying Engineering, Liaoning Technical University 
47 ZhongHua Road, Fuxin, Liaoning Province, 123000, P.R.China 
Tel: 86 418 3351520 Fax:86 418 3350478 E_mail: guijun.yang(@163.com 
Commission WG VII/3 
KEY WORDS: Land Cover, Parcel-knowledge, Automatic, Change Detection, Integrated 
ABSTRACT: Automatic land cover change detection via remotely sensed imagery has a wide application in the area of LUCC 
research, nature resource and environment monitoring and protection. Under the condition that one time (T1) data is existed land 
cover maps, and another time (T2) data is remotely sensed imagery, how to detect change automatically is still an unresolved 
problem. This paper developed a land cover parcel-knowledge guided method for automatic change detection under this situation. 
Firstly, the land cover map in T1 and remote sensing images in T2 were registered and superimposed precisely. Secondly, the 
remotely sensing knowledge database of all land cover classes was constructed based on the unchanged parcels in T1 map. Thirdly, 
guided by parcels in T1 land cover map, obtained RS image statistical information corresponding to the parcel in land cover map. 
Finally, land cover changes were detected and the change parcel was recognized through the automatic matching between the 
database of land cover classes and the extracted statistics of parcel. Experimental results and some actual applications show the 
efficiency of this method. 
1. INTRODUCE data has a great deal of land cover information consistent with 
T2. data. If the useful information is mined by the computer, 
To accurately and quickly obtain the space-temporal and the knowledge database of land cover classes based on the 
information of land cover change by Remote Sensing is the statistic information is established, the changed information 
key technology of resource and ecological environmental can be automatically and quantitatively detected via the guide 
monitoring and the international LUCC research, and it has of the parcel-knowledge. 
become the focus of the international remote sensing studies. 
Considering that multi-temporal remote sensing data is used as 2. PRINCIPLES AND TECHNICAL PROCESS 
information source for change monitoring, several methods 
have been developed, such as Image Differencing, Principal While using T1 data (land cover map) and T2 data (remote 
Component Analysis, Post-classification Comparison and sensing image) to monitor changes, there are a large number of 
Change Vector Analysis etc. to automatically detect the unchanged parcels in T1 data, which are consistent with the 
changed information! 234! However, while changes are ground surface information presented by the remote sensing 
automatically monitored, the case that time one (T1) data are image. These parcels can provide enough information. By 
land cover maps, and another time (T2) data are remotely mining the information, the effective knowledge used for 
sensed image is very common. The frequently adopted method change detection can be obtained. When T1 data and T2 data 
is registering and superimposing T1 data and T2 data, which is are registered and superimposed precisely, the region and the 
time consuming and has a big workload. In fact, the class of the unchanged land cover in T2 image partitioned by 
unchanged information is the main information; therefore T1 parcels in T1 data are easily known. 
1160