Contributors:Chao Cai, Xixia Sun, Su Pan, Zhengning Zhang, Qiyu Li
Simulation experiments results (search efficiency, region coverage rate, environment uncertainty reduction and the probability of collision of the proposed method, search methods based on the random search, greedy search, particle swarm optimization algorithm, ant colony algorithm, Q learning, conventional IWD algorithm, and the IWD algorithm with co-evolutionary strategy).
the data consists of two folders, namely Training and Test sets. each of these folders is made up of images from cameras which have been used in the analysis of the system in this article. the data is applied as training and test, and afterwards the test set is trained while the training set is tested. this is aimed at getting an average recognition rate for the system.
In this data-set, 39 different classes of plant leaf and background images are available. The data-set containing 61,486 images. We used six different augmentation techniques for increasing the data-set size. The techniques are image flipping, Gamma correction, noise injection, PCA color augmentation, rotation, and Scaling.
The classes are,
Contributors:Arthur Lima, Luiz Felipe de Queiroz Silveira, Samuel Xavier-de-Souza
Cyclostationary feature detection is one of the most common methods used in spectrum sensing. The parallel algorithm presented here can be used in multi-core processors to speedup the execution time. The implementation uses OpenMP and the Fastest Fourier Transform in the West (a efficient Fourier transform method).