CZUR Book & Document Scanner with Smart OCR for Mac and Windows

I've only had this for a day, so I may have to update this review as I use it more. Assuming something doesn't break right away, I've played with it enough to declare it VERY USEFUL, a huge leap forward in technology. Perhaps even a 'revolution' in my way of life ;) I read books from the 1800's that are long out of print, not available anywhere except maybe ONE lending library if I'm lucky. I also like to scan individual pages out of a thick music score when I'm traveling. Has worked well in both those cases on my test runs. In fact, scanning with this device has made easy what was before sometimes impossible. Seems like a great way to scan handwritten notes too, though I've not used it for that yet. What I ESPECIALLY like: - I can scan books, sheet music, etc that are lying flat on my desk. The 'autoscan' feature activates after pages have been flipped, so you can multitask and just flip pages when you're ready. It waits for you while you vacuum or whatever. I put a weight inside the 'thumb glove' to keep pages open while I'm away after flipping. Scan quality was excellent on all things I tried so far. - When not in use, I now use the scanner as THE lamp for my desk area-- it illuminates my entire work space quite evenly. They even provided 2 additional USB-rechargeable portable LED desk lamps -- a MOST pleasant surprise -- better than any I've bought on the market. - The OCR and PDF work sufficiently well for me. The page curvature elimination works very well also. The 'thumbprints' from holding the pages down sometimes don't get edited out completely; but this doesn't bother me too much. I might eventually find out what's wrong on those pages where it hasn't worked. So I gave it 5 stars despite some 'rough edges' which are as follows: - documentation is very minimal and incomplete, so you've got to figure a lot of stuff out on your own by just trying things. - installing on windows requires running the app with admin privileges, a pain. They explicitly say in the provided documentation (such as it is) that it runs only under windows not MAC, but I haven't tried obtaining MAC Software which is rumored to be available online. I have computers that run everything, so I'm OK with using it on windows for now. Linux would be my #1 choice. - GUI is very barebones and primitive, but gets the job done. I personally don't require a lot of polish in a GUI as long as I can do what I want. - When OCR-ing a book, you get blocks of text interspersed with headers, which I suppose is to be expected. I'm after text and almost never care about photos. I'm a decent python programmer and so can post-process headers out. Also, I don't like how OCR goes to *.doc files by default; I usually prefer working with straight text in this context, and the intermediate 'doc' files will require some thought for use in batch processing. All in all though, to sum up in one word: spectacular! Kudos to the developers.

This is impressive. I first wanted something like this when Google worked on their Library of Congress project around 2007 or so. I have tens of thousands of documents that I've been wanting to digitize so that I can get rid of the physical stacks. Various books too. This product absolutely does what it's supposed to do. Out of the box, this is an impressive display. I would say their packaging rivals that of Apple products (of which I also use). It's slick, solid, well-presented, and gives a good feeling of quality. I would say that hardware wise, that quality delivers. The unit is solid, easy to use, quick to assemble, and easy to store. The only thing I found a little iffy is the extra light. It's absolutely useful and while I think they did a great job with the design, the light is held in place with a magnetic rectangle lock (very similar to the power cord on a MacBook). Because the light is being pulled by gravity and is a little heavier, it can come off with just a little force. Again, not a huge deal, but it's my only complaint with the hardware. There is a foot pedal and hand pedal that allow you to scan without touching the scan area. That in itself is really convenient and functional. The software is good, but definitely leaves some features to be desired. I noticed that the Fujitsu version of this allows multiple items to be laid on the black cloth and processed individually. This device doesn't seem to allow for that. The multi-page gets off from time to time and I'll have to toy with it in those cases. As a designer and engineer, there are a number of software features that I could envision being pretty easy to build into the software. I also don't feel that the algorithm that "blends in" the thumb holders is very strong. It works, but they could definitely improve upon this significantly. I was also a little frustrated having to upgrade my version of OS X to use their scanner software. I was only half a version behind and I just can't understand what functionality of the newest OS X release they use that would warrant that. So it took me some time out of the box before I could use it since I had to upgrade my entire machine, but it's been running smoothly since. If you have a lot of documents or books that you want to scan, I would highly advise this. If you have a ton of photos, unless I'm mistaken and there IS a way to do multiple scans at once, you might consider their competitor. It's more expensive, but that one feature would save me a TON of time with my scanning process. I've had it for 2 days and have already scanned over 3,000 documents - so yea, it's pretty fast.

It works really well and it scans fast, however the processing time slows down the machine. I had to use a lower DPI and Resolution settings and not use the page flattening option and just use "original picture" or my laptop (which is not screaming fast, but a good machine that runs other high speed production level 90ppm scanners no problem) If not the laptop could not keep up with the scanning speed. it will buffer pages ahead of where you are scanning, but if you are in a production environment where speed is key the options mentioned above allow for less than 10 pages in the buffer and then you sit and wait and I mean like over 5 minutes wait before you can scan 10 more pages and then wait... not good. The page flattening option works great, but it slows down scanning a lot. If speed doesn't matter to you use it, for us it just bogged down the process way to much. Using lower dpi than desired and non page flattening allows you to be hundreds of pages out in front of the processing and then you can simply let your pc work on them while you go home or lunch or do other things. That is the only way I could make it work for us. But this technique did work and I was able to scan 4k pages or so in just a few days. At this price point this is a good scanner. You just need to play with the settings to make it work for your needs.

I downloaded the software...the online guidance from overseas is in-complete. I have had this for several months...going back several times It won't turn on and start the process...will give it another try soon, really...then if my frustration continues I'll trash it...Yeah expensive move but Instructions that ARE that vague make for mis-takes. GUESS that is why BOEING USES ENGLISH in all aspects to communicate design, production, flight training, repair, and sales... Greek in English is still GREEK. Tech writers for Cruz make assumptions in their Poor manual... I have seen pictures taken from a cruz, but how they got there is beyond me with the manuel they wrote...too bad.. Update: 09-11-2019 I have not been able to get this unit to take a picture, I don't see a book, I don't get to project a page from it, in over 2yrs. I forgot about it in it's almost pristeen boxed condition. $400 worth! Expensive Paperweight. I can't return it because it is past return time ..what I can do is buy a different product. I haven't found any I like type models of competing companies, but that's ok I've stopped trying to copy books or photo albums. My photos are taken and now scanned individually on flat beds....Thank you Canon and Epson: the photos look great, even on my VuScan portable wand and fixed units, they come out good. No messing around with foreign products with possible spyware in it either. I'm going target practice soon with my Bond Arms Old Glory Derranger with a 410 Shotgun Shell barrell, Pistol. Can't wait! to post pictures of a useless piece of garbage....BLOWN TO SMITHEREENS! NOW THATS ONE DIGITAL PICTURE FROM A NIKON D7200 I won't have to scan, because a sound trigger should fire off the camera at high speed and close focus to get pics of the Cruz parts flying away!

My wife and I have used the ET16Plus for about one month now--And we have found it very easy and fast to use. My wife scans lots of foreign language books for reference and used to use a flat bed scanner (that has a nice feature that allows books to scan only open 1/2 way--But the flat bed scanner light and mechanism would wear out after a year or two). The CZUR used LEDs (plus 3x LED lasers) and pure electronics to take a picture for scanning--Hopefully, this will last a very long time. Much faster than using the flat bed scanner and no flipping the whole book to scan each page. The side light option is very nice for scanning glossy pages (turn off overhead LEDs). Software can be installed from CD or downloaded from CZUR's website (which I did, Windows 10 laptop did not have a CD Drive). Post processing (doing OCR and combining to PDF) was very simple, easy, and fast. Have done OCR for both Chinese and English--Pretty accurate (a lot depends on image quality of source documents). I have attached both a color glossy and black&white scanned documents at the end of the review. They were simple scans and I did not take any extra care or reshoot scans (black text on dark background near corners of page can be difficult to capture cleanly unless you play around with settings). Using the software takes a bit of experimenting and understanding how it was intended to be used. For scanning, You need to select scanning (can also be used as a camera to project on a digital protector for lecture materials), then you will be taken to the post processing screen. Then select Scan. On the scan page, you need to (upper left) setup the destination directory. Upper right choose Color/B&W/etc. for your scans. Then single page or facing pages. Now you can scan (I use the supplied foot pedal). Once you are done scanning (if you make a mistake, just write down the one or two image numbers to delete) and continue scanning)., close the scan window and you will be taken back to the post processing screen. On this screen (for example), upper right "select all" images to process, hit Export and PDF+OCR) and fill in the file name (note, the PDF document does not end up in the Image directory--but up a couple levels). And it will take a minute or two to process a 20 page B&W document into your PDF+OCR. If you made a scan mistake, on the post processing screen, you can deselect your mistakes, or select the bad image and use the delete/other editing options to fix the mistakes. Note that as you go back and forth between screens, you will probably get a warning that some temporary data will be erased--So far, I have not found any downside to this message. If you scanned your data in one session (and turned off computer before post-processing), you have to open the program/scanner/scanner (scanning screen) and select the working directory (it will tell you there is existing data--just accept)--Then (without scanning) you can go back to the export screen and do your post processing as normal (or you can continue scanning if you did not complete). I am still learning about the software--But one big warning... If you have loose leaf binders to scan. Take the pages out of the binder and scan as single pages. If you try to scan "facing pages"--The software gets "confused" and does a very poor job of separating the left and right pages (the lack of solid binding seems to cause problems with loose leaf scanning). Just scan one page at a time works very well and is probably faster than trying to scan facing loose leaf pages. Unfortunately, Amazon does not let me post PDF+OCR, so I will post a few examples of scanned pages (glossy, single page B&W, and B&W two facing pages single scan).

I don't normally write reviews, but this product warrants positive feedback. It is simply one of the coolest things I have purchased online in the last few years. It totally works as advertised. I have read some of the negative reviews and I am not sure what those folks are doing that caused their respective problems. The packaging is great and the setup took me like 5 minutes (really that fast). I am using Windows 7 for this device so that may be something to pay attention too. I installed the software directly from the company's website rather than whatever was in the box. It all seems pretty simple to follow the quick setup guide and get started. For my purposes, I want a document/book scanner versus a picture scanner or something else that doesn't really seem to fit the design concept of this product. It's basically an overhead camera that takes pictures of whatever you lay down in front of it. As far as how well the product works. I scanned 7 books today in less than 4 hours. For example (I timed it because I was blown away with the speed) I scanned a 194 page book in 22 minutes. What makes this device fast is that it scans first and then after you are done with the book, you can decide to OCR to pdf or word... or just keep the images. The ocr process obviously adds some time to the overall project time, but I'd say that book was completely saved in a word file within another 10 minutes. I also used the autoscan feature and it was perfect as well. With autoscan, I did a 300 page book and the software was intuitive enough to know when I had finally turned the page and got everything ready for the next scan/picture. So at this point, I'd say this is a great product and great experience. If anything changes I'll update my review.

If you are new to digitizing a book, be aware of time use expectations: The major time consumption in digitizing is NOT the scan time, it is the cleanup time. This will depend on the quality and sophistication of the original and how good you want the book to look and what device you want to read the book on. Old books or books with lots of tables, illustrations or symbols (like math, polytonic Greek, extensive footnoting) are likely to produce marginal results that require extensive cleanup. The scanning might take 12 minutes, the cleanup might take 12 hours. If you anticipate commercial distribution, it could take several days. Your expectation and usage plans make the difference. If all you want is readability of skewed pages and no searchability on a desktop PC, CZUR will deliver quickly. Most e-ink readers like Kindle, Kobo, Nook, etc will do very poorly with a pdf file and much better with epub or mobi searchable formats. Searchable also means speakable, the text-to-speech feature of readers will function as expected only when documents are in OCR format. It is much easier (and less desirable) to achieve a non-searchable image view than a searchable OCR view depending on who will use the scanned output and how professional you want it to look. The auto-repositioning of images feature of CZUR has limited accuracy and if you want clean image based pages that look straight and well marginated to the human eye, you will need to hand crop and rotate the images. This takes time. The CSUR will correct the rotation of pages to within two or three degrees of straightness, the offset is easily visible when it is present. The type of layered pdf file where you can both see the original page image and simultaneously search the underlying textual characters is a format beneath it is not generated by the CZUR software. That layered format is the quick and dirty way that the google book project and high end unattended robotic digitizers achieve fast results. No human supervised spell check is performed and consequently hundreds, even thousands of errors remain hidden from the person reading the book. You only see those errors if you can export the document into .epub or .docx format to pass it through software with slow human mediated spell checking. OCR machine spell checking often misrenders the images it sees. The image of a distorted "O" may render as (] or (j ,may render as "O". Italicized letters like b,h, rn and cl are especially vulnerable to confused interpretation. Letters or words may be randomly italicized or bolded by mistake. The presence of page numbers or page headers/footers may interfere with your intended purpose and may have to be manually removed. An index set in extra small type (eg 6pt) may end up as mostly human readable, but chopped up in a confusing format and randomly misspelled. Many of your lower case "e" letters can turn into "c". Expectations realized with CZUR scanner: High scan rate Better than average resolution A set of JPEG files that may be loaded into other higher quality OCR software Thumbprint removal happens most of the time, but is quirky--very unpredictable and the erasure eats into the page past the actual size of the thumb image sometimes chopping into words by removing leading or trailing letters. Virtually non-existant documentation Expectations N OT realized with image CZUR output: Uniform margins Straight untilted pages that are not distracting to the eye Quality rendering of half-tone photos Automatic contrast detection may create background color drift Expectations NOT realized with OCRd CZUR output: Quality rendering of half-tone photos Rapid OCR and accurate human mediated post processing The amazon product listing mentions a bundle with ABBYY which was not present with the shipped unit. No matter how you slice it, book digitizing takes time and the better you want it to look, the more TIME it takes. In some ways the CZUR is a revolutionary packaged device for high book scan rates at a consumer-oriented price. But it does not quicken the time-consuming postprocessing aspect of digitization production. If you already own a very high-resolution 20MP camera you might achieve a similar outcome with third party software like booksorber. The absence of any meaningful documentation on such a sophisticated product is inexcusable and reason enough to return it in my view. On the other hand if your computer skills are self-taught by trial and error, this may be just the product for you. So in the end, just how fussy are you when you read the printed page, ie the distortion distraction factor? How much mental and physical energy are you willing to invest keeping things tidy? This review is based on personally digitizing several challenging books, but just doing a hardbound novel would be a piece of cake, doing poetry or a mass market paperback is a bear. My sense of this software [from a former developer] is that it is still at beta level, not enough features and definitely not up to an ADA certification audit, the type is too small and the contrast is loo low unless you have young eyeballs. To my senior eyeballs the screen has an uncomfortable washed out fuzzy look and is functionally at less than ten percent what you would find in the freebie Paint program that comes free with windows. There is a similar order of magnitude difference between the free copy of Omnipage that comes with a low end Canon flatbed scanner and the inability CZUR to allow for human mediated corrections. Sometimes the automatic image "correction" makes the image worse rather than better producing a very wavy baseline effect for lines of text. Sales claim:"Quickly book to ebook (Less than 3 mins for a 200 pages book)" After some practice, I would not attempt to scan a 200 page book in less than 30mins with this product just for the scan alone for a simple book. If you try the CZUR ET16 you will have a chance to answer such issues from personal experience. If you do decide to take the plunge, here are some tips: There are lots of youtube ET16 tutorials & tips better than the ones on the CZUR site, like those from D&H Innovation & John Willis & E-Z photo scan. If you are editing an image and want to rotate it, you can press the right and left arrow keys on the keyboard to go clockwise and counterclockwise by one degree in the rotate tool.. Make sure the book is dead center in the viewfinder, serious parallax distortions result from items placed near an edge because the camera is so close to the page. If you accidentally scan the same pages twice, you can just delete the extra images without have to renumber anything, the same is true of deleting blank pages. Some users that digitize glossy pages prefer a light shield box or shield for better results, others do their scanning in a dark room, lighting can be critical for the reflections from coated stock. Interacting with "support" is fast and easy, their reply will be "Thanks for your question, watch our videos several times, have a nice day." Assume you will end up buying either OmniPage Pro or ABBYY Pro if you expect to do any serious work, both can import folders full of images from the CZUR. Also assume you will have to make several cleanup passes for top notch results, one for human-mediated OCR, one for duplicate or missing pages, one for a word processor spell check, one for page aesthetics including margination and justification, and a set of searches of OCR misread trigger words like arid,arc, riot, sonic, clay, yon, ina, docs, tip, nay, and hut. And of course the final test is actually reading the book for ugly surprises, even if all you do is read one or two chapters you will learn funky things that will repeat all through the book that you can search for. As pretty as the packaging is, the bottom line costwise is that this item will set you back about twice what I am willing to pay considering the very low cost of a Sony 20 mp cameras and the weak software. The unit was returned until the price/product usability becomes more realistic. Other options are way more attractive to me. A user friendly product would require a total rewrite and a clear, complete manual.

After reading the reviews of the Fujitsu scansnap sv600, I decided to give this Czur scanner a try and I absolutely like it. Yes, it’s from China and the user manual is not in proper English, but the setup process was very simple and straight forward. The scanning software is also very simple with only three main operations, Scan, Archive, and Bulk, which are used in scanning the page, running OCR and processing the scanned ages in bulk. Without disassemble or roughly handle the book, scanning a thick book is just a matter of turning the pages with the book opened facing up, and the built-in software will correct and straighten the curved lines automatically, the result was pretty amazing. Compared to scan one page at a time facing down on a flatbed copy machine, this little handy scanner does the job much faster and easier and best of all there is no need to do any manual editing. So far I managed to scan over a dozen books without any problem. I haven’t tried the cloud scanning option yet, but already felt my money was well spent. If anyone who needs to do some serious scanning, the Czur scanner is definitely a good choice and I highly recommend it.