Outputs

D.T2.3.1- Information Technology (IT) platform for antifraud customs controls

The projects ISACC took care of the development of the platform, including all necessary adaptations, also deriving from the Customs agencies feedbacks, until the very end of the project. One of the main purposes of the platform is to make a comparison between multiple Custom Footprints (CF) that contain different types of data such as numeric, text, and imagery and generate a response with the differences and similarities between them.

It integrates data, coming from heterogeneous technologies and systems, in order to provide a rich information base supporting customs authorities during antifraud controls. It uses the SIFT algorithm, for the automatic detection of visual changes between scanner X-Ray images.

Credentials are differentiated per Country and per priviledges (for admin, single operators etc.)
After the pre-processing phase the couple of images has the same scale and rotation, it’s possible to actually perform their comparison. The Feature Matching algorithms can also be used to find as many common graphic features as possible between the images, and use the spatial sparsity of the found common descriptors as an indicator of the presence (or absence) of differences between the two images.
The high density of descriptors found in a certain area indicates absence of differences and hence homogeneity. On the other hand, low density or absence of descriptors indicates a high probability of differences which may lie in that particular region.

The image comparison algorithm, as mentioned, is based on the SIFT algorithm: the algorithm can be divided in three steps:
1. keypoints and descriptors identification
2. descriptors matching
3. similarity percentage calculation.

The application has several components each of which have different functionalities:
- Create Users
- Manage Users
- Insert a specific type of data (Custom Footprint)
- Upload multiple data
- Compare the content of data and generate a response
- Compare images between them and generate a response
- Display the route it took for the file to arrive from point A to point B
- Add validation on operation based on user logged in.