This shows you the differences between two versions of the page.

Link to this comparison view

engine:on_demand:description [2018/02/01 17:21]
pdufour [Administration Console]
engine:on_demand:description [2018/02/01 18:15]
Line 1: Line 1:
-===== Hosted Software Platform ===== 
-LTU engine/ON demand is a hosted software platform that provides access to JASTEC’s comprehensive image recognition functionalities. The LTU engine/ON demand platform is highly flexible and is designed to adjust and scale with changing demands. ​ 
-A single account for LTU engine/ON demand can be used to add image recognition functionality to multiple applications. 
-===== Administration Console ===== 
-The LTU engine/ON demand Administrative Console is accessed via. 
-URL: https://​www.ltu-engine.com/ ​ 
-See the Administrative Interface section below. 
-**LTU engine/ON demand** has the same Administrative Interface as the licensed version of LTU Tech software: ​ **LTU engine/​server**. ​   
-Some aspects unique to LTU engine/ON demand found in comments marked with this symbol: ​ :!:  
-===== Application Programmming Interfaces ===== 
-JSON API:  the JSON API is the recommended API for new applications. ​ The licensed version of LTU engine includes the JSON API, which facilitates the porting of applications from LTU engine/ON demand to LTU engine/​server. ​ Query examples in the JSON API documentation generally reference LTU engine/​server.  ​ 
-:!: LTU engine/ON demand users will know to adjust the query examples to the proper /ON demand references. ​ 
-===== Integration Check List ===== 
-Below is a quick checklist to help prepare for the integration of LTU engine/ON demand. ​ 
-  ***Signature type** ​  - What feature(s) needed? (similarity,​ matching, color) ​ 
-  ***Storage** ​ 
-    *Estimate the number of images that will be included for the application ​ 
-    *Estimate the image volume changes per month on this application ​ 
-  ***Search** ​ 
-    *Estimate number of searches per month  
-  ***Development** ​ 
-    *List the public IP addresses of all machines that will make direct calls to the API interface. ​ 
-    *Launch sample code provided by LTU Tech to check that everything works  
-    *Start development and unit test  
-    *Validate (pay attention at the refresh delay per application) ​ 
-    *Empty the database and start loading content for the application ​ 
-  ***Production** ​ 
-    *Monitor the number of queries per month (also available on the LTU engine/ON demand administrative console) ​ 
-    *Continuously evaluate the capacity used monthly in terms of queries and image database volume ​