This usually occurs when you type something wrong in the activation code (e.g. problem with the case, missing number or letter or an additional blank after the code). Try again to copy and paste your activation code because it is case sensitive.
Please try again. As the activation code is case sensitive, it is usually better to copy/paste it. Please be careful not to enter leading or trailing spaces.
For image files: You will see the words "Unregistered copy" written in white on the output image made with the demo version.
For text files: you can't export OCR results but you can preview them within Prizmo.
The demo version enables you to try the application before actually buying it. Moreover, there's neither use limit nor time limit for trying the demo version.
No, it isn't. Prizmo relies on some technologies only available in Leopard. It would be very difficult and time-consuming to develop exactly the same application compatible with Tiger. And the ouput images wouldn't be as good as they are under Leopard.
Creaceed is specialised in Mac and iPhone development. We started 10 years ago to work with all Mac-related technologies (Cocoa, Objective-C, OpenGL...), and we think they are great. So, we want to go on this way. Sorry for Windows users.
Open the Preview application. Click File, and then Import from Scanner. If your device is listed there, it is supported by Snow Leopard.
And don't forget Prizmo's trial version is fully functional. So, you can also check if your scanner is supported by importing a document into Prizmo right from your device.
We rely on the tethering function as implemented in Mac OS X Snow Leopard, and Apple does not provide an official list of supported devices. However, we think that Nikon and Canon DSLRs are a safe bet.
You can rotate the selection grid just the way you want. The arrow next to the grid shows the upper side of the output image. There are 2 ways to rotate the selection grid :
click Tools > Rotate Grid Left or Rotate Grid Right.
or press simultaneously Command + L + ⌥ to rotate left, or Command + R + ⌥ to rotate right.
In the lower right corner of the left part, you can move a slider to set the zoom level applied on the input picture in the right part of the screen. Besides, there are several zoom options available in the Tools menu:
Zoom in or Command and + keys
Zoom out or Command and - keys
Zoom to Fit Grid or F key: with a single click, you'll see only the selection grid.
Zoom All or A key: with a single click, you'll see the whole input photograph.
Moreover, you can move the input picture in the right part of the screen: click the picture, hold and move until the picture is where you want it to be.
in an approximate way: when you mouse over the output preview, white lines appear on the image sides. Click one of these lines, hold and move the mouse. The output image is resizing in real time.
in a very accurate way: show the Page Dimensions pane (in the Tools menu). Then, choose a standard preset like US Letter or opt for a custom format. For the latter, you can choose between pixel, inch, centimeter or millimeter, and then type a number in the fields "Width" and "Height".
Yes, you can if you use a tripod and if you keep the book in the same position when you take pictures. Here is an easy way to automate image processing within Prizmo:
For the 1st picture of the series, create presets in the image processing pane by clicking the + button.
As from the 2nd picture, import the photograph, and choose the Last Grid option to reuse the same extraction grid across shots.
Prizmo’s purpose is to correct the distortion of a single image using the data coming from camera calibration. And Prizmo’s calibration assistant is for standard camera images only.
However, you can try to get panoramic images straight by manually tuning the 1st and 2nd Radial distortion settings in the Lens Options pane.
We checked the accuracy of the OCR engine in the 10 available languages. Even with pictures taken in poor circumstances (document on several other documents, picture taken in front of the document rather than above...), we got very good results.
Here are some recommendations for improved OCR results:
you should be right above the document you want to photograph.
use the camera flash when you take a picture.
when you locate the selection grid, check the grid is larger than the text to be recognized.
make sure the top and bottom lines of the selection grid are parallel to the text to be recognized.
choose the language of the document in the list (you can recognize a letter in Spanish while using Prizmo in English).
zoom in to get the text area in close-up, and play with the binarization slider. You should be able to read the text areas on the whole page but don't binarize too much (or you would get worse results).
Image quality is a major issue indeed. OCR results will most likely be disappointing if the input image quality is poor. Here are a few examples of what should be avoided:
The image is blurry. Check you are able to read the text on the picture. You can't in this example. Please take a new picture of better quality.
The ticket is crumbled.
The image below is a screenshot, and not a picture taken with a camera. The issue is the resolution difference (72 dpi for a screenshot, and 300 dpi for a picture) which has an impact on OCR results. It is much more difficult to recognize small characters on a screenshot than on a photograph.
It is a double-sided document. Characters, lines and colors which are printed on the other side could hinder OCR.
No, we aren't. Prizmo's OCR engine supports 10 languages, and some of them are actually widely used all over the world (Spanish, English, French, German...).
Besides the usual image files (jpeg, tiff and png), there are several text files available:
PDF (Text): text is formated (columns, character size, bold, italic...), and it creates a new clean document.
PDF (Image + Hidden Text): it is made of formated text (columns, character size, bold, italic...) in addition to an image of the input document. If there are mistakes in the OCR-recognized text, you can correct them thanks to the image. It's the best file type for archiving purposes.
RTF: text is partially formated (only bold, italic...). Very handy if you want to copy/paste text in Pages, Mail, OpenOffice...
TXT: it's raw text without any formating. For people who want to modify the lay-out of an existing text.
No, it isn't. Before calibrating your camera, check if a calibration model is already available on Creaceed's servers. Many users have already sent their calibration results to Prizmo's database, and they are shared with the community.
Import your picture into Prizmo, open the Lens Options pane, and click Search.
A model is available: the Camera Optical Data box above the Search button contains some information about your camera, lens and zoom. Great! You don't actually need to calibrate your camera (but you can if you really want to).
No model is available: a pop-up window warns that no model is available ("Optics not found"), and asks you either to use Prizmo without correcting lens distortion, or to calibrate your camera. We advise you to take a few minutes calibrating your camera as it will help Prizmo correct lens distortion more efficiently. You can notice lens distortion on objects near picture's sides. For instance: a wall or a tree will look like curved, rather than straight.
It seems that Prizmo didn't manage to locate the pattern on the pictures. There are several rules to follow when you take pictures of the pattern if you want to create a reliable calibration model.
the pattern should be as big as possible (taking a picture of the pattern at 6 feet from the flat screen is useless)
the whole pattern should be on the picture with some white space all around
you should take pictures from different points of view.
Basically, you have to calibrate each camera body + lens + focal. Once calibrated, you can import a picture with the same attributes, and Prizmo will automatically select the related calibration model. You can see which model is being used in the Lens Options pane (the <1/n> text next to the Search button).
If you can't see that <1/n> text in the Lens Options pane,
either none of your calibrations (camera body + lens + focal) can be used for the imported picture
or the picture you imported misses some EXIF info (this can happen with older lenses).
Remark: Be careful with varying focal lenses, like a 16-35 mm: if you calibrate the 16 mm focal, it will work only with 16 mm photographs. If you set the zoom to 20 mm instead, the calibration won't fit. You’ll have to calibrate camera body + lens + 20 mm focal as well.
