Convoy Analysis

Convoy Analysis allows you to search for vehicle co-occurrences. You can detect one or more unknown vehicles traveling in convoy with known target vehicles. This can help detect:

vehicles following cash in transit vehicles
vehicles that were previously unknown, but have been identified traveling with known or suspected criminals
vehicles traveling with stolen vehicles that are subsequently used in criminal activity
compromised covert operations

Source target vehicles can be entered individually or you can use a list stored in a text file (perhaps generated as a result of Geographical Profiling for example).

Select the date and time boundaries for the Plate Analysis records to be searched. With this done, the time between the target vehicle passing a camera and any other co-occurring vehicles passing the same camera can be set in the time window along with the minimum number of co-occurrences. The results can be further restricted by choosing to return only the vehicles that co-occur most frequently.

Note: If you are looking for co-occurrences on a particular day, you could retrieve records for co-occurring vehicles that were actually recorded on a preceding or succeeding day. The time you set relates only to the target vehicle. For example, if a target vehicle is recorded at one minute to midnight, and the time window is specified as five minutes, you could detect co-occurring vehicles at up to four minutes past midnight.

The target vehicles and their co-occurrences are clearly displayed and any co-occurrence can be selected for a more detailed view by simply double clicking the item.

Convoy Analysis has a number of screens, click Next or Back to move between them.

Specify Parameters

Enter the VRMs of the target vehicles that you want to analyze; they can be entered individually or you can use a list stored in a text file (perhaps generated as a result of Geographical Profiling for example).

Enter the VRMs in the Target VRMs box, one VRM per line separated by a carriage return. Alternatively click the Browse button to retrieve VRMs from a text file.

Select the date and time boundaries for the Plate Analysis records to be searched. The date and time range relates to the times that target vehicles are recorded in an ALPR read. Set a start and end date and time.

If you only want to consider a certain period each day, for example you may want to analyze Plate Analysis records between 09:00 and 09:30 each day, turn on the Include records only within these times check box and set the From and To times below.

Finally set the following options:

Time Window: Set the time window that defines a co-occurrence. For example, if you set a time window of 10 seconds (the default), it means that vehicles recorded as passing an ALPR camera within plus or minus 10 seconds of a target vehicle are considered to be in convoy. A range of values, up to 5 minutes is available. The larger the time window, the more results you are likely to get.
Minimum Co-occurrences: Set the minimum number of co-occurrences that you want to consider for each target vehicle, you can set any number between 1 and 10. For example, setting 2 (the default) means that you only want results where the vehicle has been in convoy with a target vehicle on more than one occasion. Setting a higher number could reduce the number of results.
Ranked VRMs: You can choose to see all results by selecting All, or you can specify a Number of results to see. For example if you set the number to 10, you will only see the top ten ranked co-occurring vehicles, sorted by the number of co-occurrences.

Examine Co-Occurrences

This screen allows you to see how many co-occurrences have been detected; each co-occurring vehicle is listed along with the number of co-occurrences. If you have no results, or too many, you can go back and change the parameters by clicking the Back button.

You can see the Plate Analysis records relating to the convoy by selecting one or more groups of results, and clicking the Show Records button.

Note: If you select a result with five co-occurrences for example, when you show the records you will see ten records; for each co-occurrence you will see the Plate Analysis records for both the target and co-occurring vehicle.

Select one or more of the co-occurrences that you want to analyze in the grid view. When you have selected the results, click Next to go to the grid view.

Create a Grid Display

Co-occurrences for each target vehicle are displayed on a grid. The best way to understand the display is to interpret the example shown above. There are three target vehicles; each target has a page showing all the co-occurrences that relate to that vehicle. The display shown above relates to the page for VRM: P2 DCM.

The first column of the grid shows the date and time that the target vehicle was recorded by an ALPR camera (this only shows the times that the target vehicle was recorded as a part of a co-occurrence, the vehicle may have been detected on many other occasions, but only co-occurrences are shown here).

The second column, headed with the target vehicle VRM, shows the name of the camera that recorded the read and shows the time that the target vehicle passed the camera.

The third column, headed with the VRM: MJ75 IRV, shows the camera name that recorded a co-occurrence and the time (in seconds) relative to the target vehicle. The first entry in the third column shows that MJ75 IRV passed Camera 1 one second after the target vehicle (if the time had been shown as -1, it would show that the target vehicle was recorded after MJ75 IRV).

Each co-occurring vehicle has its own column; those shown with a bold heading, F576 SQZ and HH18 CIK in the example above, are themselves target vehicles; you can see that they have their own pages.

You can copy the results for the current grid to the clipboard by clicking the Copy to clipboard button. Alternatively you can copy the results for all grids by turning on the Copy all grids check box before clicking the Copy to clipboard button.

From the clipboard, you can paste the results into another application such as a spreadsheet or text editor.