This software package contains the code for the work presented in [1] that has 5 different feature extraction and selection methods, 2 of which are novel.

These methods are as follows :

(1) CLASSICAL SELECTION (CS) (2) REVERSED SELECTION (RS) (PROPOSED-1) (3) PATCH-BASED SELECTION (PB) (PROPOSED-2) (4) LOCAL BINARY PATTERN (LBP) (5) HISTOGRAM OF GRADIENT (HOG)

Each folder for the corresponding method also contains required INFRARED (IR) and VISIBLE LIGHT (VL) image databases to be used for testing.

Image set contains 20 IR and 20 VL ones, 10 of IR and 10 of VL contain power line while the rest do not.

To run each method, executing the mainNAME-OF-METHOD.m file in each folder is sufficient. We also have a README file per method in each folder.

[1] YETGIN,O.E.,GEREK,O.N.,AUTOMATIC RECOGNITION OF SCENES WITH POWER LINE WIRES IN REAL LIFE AERIAL IMAGES USING DCT-BASED FEATURES, Journal of Digital Signal Processing, 2017.