Use Tools»Batch Training to train a folder of images. Complete the following steps to train a folder of images:
- Select Tools»Batch Training.
- Click Browse to open the Select a folder of images dialog box.
- Locate the folder that contains the images which you want to batch process. Click Open to open the folder, then click Current Folder to set the folder as the image source.
Note The Batch Training tool will only process the selected folder, and will ignore the contents of any subfolders.
- Click Draw ROI and draw an ROI around the sample that you want to train. Ensure that the ROI will encompass the sample to train in all images in the folder and click OK.
Note If you do not draw an ROI, the largest sample will be trained.
- Choose one of the following methods to specify a label for each sample:
- If all images in the folder are of the same class, enable the Use the same label for all the samples option and enter the label in the Class Label control.
- If the images are not of the same class, you must use a labels file to specify individual labels for each image.
- Enable the Use a file that specifies the labels of the samples option.
- Click Browse to select a Labels File Path. Each line of a labels file lists an image name followed by the class name assigned to the image. A delimiter separates the image name from its label.
Note If an image is not listed in the labels file, it will not be processed.
- In the Delimiter list, choose the correct delimiter for the labels file. If an appropriate delimiter is not listed, choose User Specified in the Delimiter list and type the appropriate delimiter into the User Delimiter field.
- Click Train to process all of the images in the selected folder. The training process both trains and classifies the samples.
Once processing is complete, the Batch Results dialog box will open.