≪ View Page

Diferenzas

Isto amosa as diferenzas entre a revisión seleccionada e a versión actual da páxina.

Enlazar a esta vista de comparación

Both sides previous revisionPrevious revision
Next revision		Previous revision
Next revisionBoth sides next revision
hiperespectral:sae-cd [2018/01/16 17:38] – [Downloads] javier.lopez.fandinohiperespectral:sae-cd [2018/01/16 17:53] – [Outputs] javier.lopez.fandino
Liña 32: Liña 32:
   * Training samples randomly chosen in each run.    * Training samples randomly chosen in each run. 
   * 10 independent runs for each classifier.   * 10 independent runs for each classifier.
-  * SVM classification carried out using the LIB-SVM library and the Gaussian radial basis function (RBF) +  * SVM classification carried out using the LIB-SVM library and the Gaussian radial basis function (RBF). 
   * ELM configured with a sigmoidal activation function.   * ELM configured with a sigmoidal activation function.
  
Liña 38: Liña 38:
  
 === Image files === === Image files ===
-{{:hiperespectral:referencedatacolorhermiston5.png?200|}} +|Reference data of changes |Binary CD map |Multiclass CD map| 
-{{:hiperespectral:binarycd.png?200|}} +|{{:hiperespectral:referencedatacolorhermiston5.png?200|}}|{{:hiperespectral:binarycd.png?200|}}|{{:hiperespectral:svmcolorhermiston.png?200|}}|
-{{:hiperespectral:svmcolorhermiston.png?200|}}+
  
-=== Accuracy results === 
-Binary CD accuracies.  
-|  Corect | Missed Alarms| False Alarms | Total Error| 
-|---|---|---|---| 
-|  77020 (98.74%) | 509 | 471 | 980 (1.25%) | 
  
  
 +=== Accuracy results ===
 +==Binary CD accuracies==
 +|Corect |Missed Alarms|False Alarms |Total Error|
 +|77020 (98.74%) |509 |471 |980 (1.25%) |
  
-Multiclass CD accuracies. 
  
  
-|  Classifier | Parameters   |  FE | OA (%)| AA (%)| Kappa| +==Multiclass CD accuracies== 
-|---|---|---|---|---|---| +|**Classifier**   | **Parameters**     |**FE**   | **OA (%)**  | **AA (%)**  | **Kappa**  | 
-| ELM | N=120  | PCA  | 91.73  | 76.06  |  86.83 | +| ELM             | N=120              | PCA     | 91.73       | 76.06       | 86.83      | 
-| ELM | N=120  | NWFE |  91.76 |  76.75 |  86.83 | +| ELM             | N=120              | NWFE    | 91.76       | 76.75       | 86.83      | 
-| ELM | N=60  | SAE  | 95.19  | 90.45  | 92.31  | +| ELM             | N=60               | SAE     | 95.19       | 90.45       | 92.31      | 
-| SVM | C: 64.0 γ: 32.0  | PCA  | 91.46  | 71.16  | 86.46  | +| SVM             | C: 64.0 γ: 32.0    | PCA     | 91.46       | 71.16       | 86.46      | 
-| SVM | C: 32.0 γ: 16.0   | NWFE  | 91.29  | 90.61  |  86.05 | +| SVM             | C: 32.0 γ: 16.0    | NWFE    | 91.29       | 90.61       | 86.05      | 
-| SVM | C: 32.0 γ: 0.0625  | SAE  |  95.52 | 92.56  | 92.90  |+| SVM             | C: 32.0 γ: 0.0625  | SAE     | 95.52       | 92.56       | 92.90      |
  
 C: penalty term in the training of the SVM. γ: radius of the gaussian function of the SVM. N: Number of neurons in the hidden layer of the ELM. FE: Feature Extraction method. C: penalty term in the training of the SVM. γ: radius of the gaussian function of the SVM. N: Number of neurons in the hidden layer of the ELM. FE: Feature Extraction method.