Classes | |
| class | AbnormalChange |
| Looks at refmax of successive elevation scans and alerts when the change in maximum is abnormal. More... | |
| class | AdaptableParameters |
| Contains all the adaptable parameters for QCNN. More... | |
| class | BlobSegmentation |
| class | BloomPreproc |
| Maintains information about bloom using the hybrid scan. More... | |
| class | CloudCoverNN |
| An extension of the radar-only neural network, this also uses the cloud cover field. More... | |
| class | CloudCoverProvider |
| Will provide a radar specific cloud-cover indexed radial set. More... | |
| class | Entropy |
| Calculates the information content (entropy). More... | |
| class | PolarImageUtils |
| class | Postprocessor |
| Base-class of options for post-processing the QC field. More... | |
| class | NullPostprocessor |
| Does no postprocessing. More... | |
| class | CombineMedian |
| postprocesses the precip confidence field by computing a median. More... | |
| class | CombineMean |
| postprocesses the precip confidence field by computing a local mean. More... | |
| class | CombineSmartMean |
| postprocesses the precip confidence field by computing a local mean, but taking into account the distribution of high confidences in either direction. More... | |
| class | CombineProbabilities |
| Uses a joint probability function of the local pixel-wise probability values. More... | |
| class | CombineMaximum |
| The most conservative method: a local maximum. More... | |
| class | QCNN |
| Neural network to perform quality control of radar reflectivity data. More... | |
| class | QCTrainedNetwork |
| Uses the trained neural network to return probability of precip. More... | |
| class | RadarQualityControl |
| Abstract base class of objects that perform quality-control on reflectivity data. More... | |
| class | RadialPreproc |
| Maintains information about bad radials in a virtual volume. More... | |
| class | Statistics |
| Abstract base-class of all the image statistics classes. More... | |
| class | BasicStatistics |
| class | LocalStatistics |
| Statistics computed in a local, polar, neighborhood. More... | |
| class | BlobTextureStatistics |
| class | VerticalStatistics |
| class | CompositeStatistics |
| Comprised of a number of individual Statistics. More... | |
Functions | |
| Length | getGateWidth (const PolarGrid &pg) |
| Length | getGateWidth (const RadialSet &rs) |
| template<class RSorPG> | |
| void | setTerrainBlockage (const std::string &terrainFileOrDir, int maxGates, const RSorPG &pg, RadarQualityControl *qcnn) |
| In NN training, we might run qcnn on a set of cases from different radars, so this function supports that by reading from a directory with like-names if needed. | |
| Length w2qcref::getGateWidth | ( | const PolarGrid & | pg | ) |
| Length w2qcref::getGateWidth | ( | const RadialSet & | rs | ) |
| void w2qcref::setTerrainBlockage | ( | const std::string & | terrainFileOrDir, | |
| int | maxGates, | |||
| const RSorPG & | pg, | |||
| RadarQualityControl * | qcnn | |||
| ) |
In NN training, we might run qcnn on a set of cases from different radars, so this function supports that by reading from a directory with like-names if needed.
If a directory is passed in, the terrain file should be named dir/KINX.nc etc. i.e. using the 4-letter identifier.
The terrain file is used to correct the VerticalStatistics.
1.4.7