Internet has been employed as an efficient avenue for the dissemination of a variety of geospatial information. However, the distribution of the geospatial processing capabilities over the Internet, especially those related to geospatial image processing, is still lacking. Many customized solutions based on CORBA (Component Object Request Broker Architecture) and DCOM (Distributed Component Object Model) have been developed to deliver geoprocessing functionality online but they are often designed to operate on specific datasets and they cannot interact with other online data sources easily. Geoprocessing is usually achieved through a series of steps. The inability of these solutions to support generic online data makes it impossible to chain with other systems to perform sequential processing. The inseparability of data and geoprocessing functionality of these online solutions has limited their usability.To achieve the separation between online data suppliers and geospatial processing functionality, a framework of geoprocessing web services is proposed in this research. An online image geometric rectification web service is developed as an example. This web service allows the images to be rectified and transformed into a known geographic projection (with map coordinates). The reference images or maps with known projection can be obtained from a data supplier web service. The users of the client application can identify ground control points (GCPs) from both the reference digital map/image and the local input image. The proposed image geometric rectification web service then performs the actual rectification based on these GCPs and returns to the client application the transformation parameters estimated through a least square solution. Due to the use of industry standard technologies such like XML, SOAP, WSDL, and UDDI to build geoprocessing web services, it is possible to distribute the geoprocessing capabilities in a platform and language independent manner over the Internet, with faster client application development, easier service deployment and efficient online discovery.