ROBIN Competition

Datasets, Ground Truths and Metrics
For the evaluation of
object recognition and image categorisation
algorithms

THALES (dataset #6)

Image categorisation produced by THALES.

This dataset contains scenes observed from two airborne infrared sensors, sampled along a sequence of views under various viewing conditions and environements.

The sensors used are

  • Micro Bolometer, generating images of size 320 x 240
  • InSb sensor, generating images of size 320 x 256
  • QWIP sensor, generating images of size 640 x 512

The THALES database contains 8 bits jpg and 16 bits tiff images, depending on the sensor and compression.

Classification Tasks

IN : list of classes to research, test images with list of coordinates of object centers.
OUT : detected class of object for each object center.

C1 : Discrimination of Vehicles : Car, Truck, Bus.

THALES_C1_1.png THALES_C1_2.png

C2 : Discrimination of Cars : Halted Car, Moving Car, Parked Car.

THALES_C2_1.png THALES_C2_2.png

Detection Tasks

IN : list of classes to research, test images.
OUT : list of detections (bounding boxes or point) for each image.

D1 : Detection of Boats

THALES_D1_1.png THALES_D1_2.png

D2 : Detection of Cars

THALES_D2_1.png THALES_D2_2.png

D3 : Detection of Vehicles : Car, Truck, Bus.

THALES_D3_1.png THALES_D3_2.png