Implémentation du logiciel CODE en C++ – Comparatif des performances et des résultats


  1. Jobin, P.M.
Corporate Authors
Defence Research and Development Canada, Valcartier Research Centre, Quebec QC (CAN);Pierre-Marc Jobin, Levis Que (CAN)
This report describes the C/C++ implementation of the performance prediction software CODE (Correction and Detection Evaluation). The software will be usable by the client to plan his missions according to the terrain and the weather conditions in which a sensor operating in the LWIR (Long Wave InfraRed) band of thermal infrared hyperspectral could be deployed. To estimate the detection performance the software use an already corrected hyperspectral image; the sensor specifications; the weather conditions; various atmospheric correction algorithms; various targets’ characteristics and various detection algorithms. The user can specify any of these parameters according to the operational context. For example, the atmospheric profiles and the cloud coverage corresponding to present conditions could be used to compute the degradation introduced by the atmosphere on the pixels of the image. The client will also be able to select the emissivity that he wants to detect. The performances emitted by the software are ROC (Receiver Operating Characteristics) curves. These curves are plots of the probability of detection as function of probability of false alarms. At first they provide a method to select optimal detection thresholds according to the circumstances. In a second time they provide a mean to evaluate the feasibility of a mission depending on the operational conditions and for example to determine if the false alarm probabilities are sufficiently low for a given probability

Il y a un résumé en français ici.

Hyperspectral Imagery;Target Detection;Tracking and Multi-Sensor Fusion;Atmospheric correction
Report Number
DRDC-RDDC-2018-C219 — Contract Report
Date of publication
01 Sep 2018
Number of Pages
Electronic Document(PDF)

Permanent link

Document 1 of 1

Date modified: