This tool extends the third-party already installed PDF preview handler (the shell extension used to show a read-only preview of the file's contents in the view's reading pane) with extra functionality, namely metadata edit and capture, easy switching of the active preview handler and possibility to have the previewer showing in a stand-alone window (handy for multi-monitor systems).
The tool registers itself as the system PDF preview handler and proxies the real preview handler. It searches the system for all the installed PDF preview handlers, and even full PDF readers compatible to use as previewers, enabling the user to easily switch, among the valid previewers found, at any time. This is important depending on the task being performed. For metadata capture operations a previewer with text selection and copying functionality is handy, but a simple previewer is much more fast just for previewing the PDF.
The bottom metadata pane shows all the file metadata properties set to be visible in the Windows Explorer details pane (this list can be defined from the manager property handler settings tool) and enables the edition of the properties that are writable. Having this functionality, that is similar to the provided by the Windows Explorer details pane, directly controllable from the preview pane is important for the metadata capture operation, as it is not technically possibly (at least easily) to interface with this Windows Explorer details pane. In Windows 8.x it's not even possible to have the preview pane showing at the same time as the details pane!
To assist the metadata capture operation, this tool has a screen snapshot tool, an image OCR tool and a clipboard monitor. The snapshot tool, fired from the top toolbar snapshot button, permits the selection and capture of a rectangular area of any part of the screen, but it's aimed to capture a snapshot of the text rendering we want to extract from the PDF. This tool is needed because most of the preview handlers don't have text extraction functionality, as they merely render the PDF as an image.
To convert this snapshot to text, the tools have an image OCR (optical character recognition) functionality, that pops up immediately after the snapshot is taken (if the main menu enable the metadata capture assistant item is checked).
For good OCR results, it is important to take the snapshot with good text rendering quality. The previewer should be resized, or the PDF page zoomed, to have well defined text in the area being captured. The OCR tool scale option, that resizes the captured image, can be used to further enhance the text visual quality, that will determine the quality of the OCR operation.
The thresholding method settings define how the conversion, to black and white, of the captured image is done (the OCR operation works with black and white images). The global method uses a fixed gray color level to set all the colors, below this level, to black and all the rest to white. This works fine for text on simple background colors, but fails if the page background is more complex. The adaptive method should be used in this more complex page background scenarios. This alternative method divides the image in small areas and applies an optimal threshold level, to each, independently.
The selected thresholding method settings only need to be manually tweaked if the OCR operation result is bad. The OCR works best with well-defined text, and the manual tweak of this settings usually enhance the results.
It is also important to have the text language option equal to the language of the text being recognized, as the operation uses dictionary statistical methods to validate the word candidates.
The OCR operation is initiated clicking the OCR named button. After the OCR operation ends, the recognized text shows in the tool text tab. It is also this tab that shows up if plain text, not image, is copied to the the Windows clipboard, for example when using a preview handler with text selection and copy capability.
From this text tab it is possible to format and edit the text, and send it directly to one of the metadata fields using the corresponding list of buttons at the bottom. This buttons are only visible if the metadata pane is also visible, by enabling the main menu view the metadata pane item.
The metadata can also be edited manually from this metadata pane. The edited metadata is not automatically saved. The user must expressly press the ENTER key (while not editing a field), or the corresponding save button, that show at the right of the metadata pane when changes are available. While in this same, changes exist, situation, pressing the ESC key, or the corresponding button, cancels all, not yet saved, changes.