JDO - JPEG Disk Optimizer 4.0 A unix command prompt driven program for optimizing and watermarking multiple JPEGs, GIFs, and PNGs on a disk or file server. JPEG Disk Optimizer is designed to optimize multiple JPEGs, GIFs, and PNGs in a directory and, optionally, in sub directories. JPEG Disk Optimizer is a command prompt program and so can be easily driven from other programs or scripting languages. JPEG Disk Optimizer includes MagiCompressionTM 2 to reduce the size of JPEGs by up to 50% more than standard JPEGs. This can dramatically cut the cost of disk storage, bandwidth usage, server load as well as speeding up web pages. Version 2 adds the ability to resize the images and to make thumbnails and HTML index files automatically. Version 2.5 adds the ability to more finely control the amount by which JPEGs are recompressed. Progressive JPEGs are now supported. Magicompression 2 has also been added. You may find you can increase MagiCompression to 80 or 90%. MagiCompression 2 works well even if the original image is noisy or blurred. In fact you may find that MagiCompression actually improves some images. Version 2.6 adds improved MagiCompression (images are smaller and less likely to go 8x8 blocky) and an advanced de-spot and de-noise algorithm. Version 3 adds Digimarc® invisible watermarking (Windows only). N.B. Version 3 uses many new and different switches. Version 2 commands and scripts are not necessarily compatible with version 3. Version 3.10 adds visible watermarking i.e. the ability to add an image on top of the main image and to add text to the images (Windows only). Version 3.30 adds alpha support to visible watermarking and advanced noise reduction. Version 3.34 adds the ability to write the files to a different directory (-op) (Windows only). Version 3.36 adds the ability to append text to each file (-oa). Version 3.41 adds the ability to change image resolution - DPI (-od) Version 4.00 adds GIF and PNG compression, image enhancement (-l), crop (-rc) and compress to a given number of bytes (-qb). Unix versions of JDO are available. Note the functionality is slightly less. Warning: JPEG Disk Optimizer is designed to replace all the JPEG files it finds in most modes. Please ensure you have a backup or master copy of your images. Example: jdo -q 80 -ss *.jpg Compress all JPEGs with quality level 80. Write a summary of space saved at the end. nice find /tmp/images -type f -iname '*.jpg' -exec jdo -q 90 -m 40 {} \; Compress all JPEGs in the directory and sub directories of /tmp/images. (see below for more details) Overview: JPEG Disk Optimizer is designed to compress multiple JPEGs in a directory and, optionally, in sub directories. JPEG Disk Optimizer is a command prompt program and so can be easily driven from other programs eg perl scripts. Installation: Copy jdo into a directory in your path. Detailed usage: jdo [-Switch [Arg]] file(s) General Options -h[nn] Display help screen(s). -h for all screens, -h0 for summary screen, -hnn for specific help. -n Don't re-compress the file. This flag will instruct JPEG Disk Optimizer to write a marker into the JPEG file. If you then attempt to compress the file again the marker will be detected and the file skipped. Note that for this to work the -n flag must always be specified. -e Stop on error. Normally JPEG Disk Optimizer will carry on compressing files even if it encounters a problem with a file. -a filename Advanced options config file (see below) JPEG Compression -q nnn Set output quality 1-100 (1=low, 100=high). Note, if you do not specify a quality, the quality of the input JPEG will be used (see also -a switch). You can also use -q low, -q medium, -q high. -m nnn Set % of MagiCompression, 0 = none, 100 = max (Default = 0%) (see also -a switch). You can also use -m low, -m medium, -m high. -qb nnn Set maximum size of image in bytes. Quality etc will be reduced until the image is less than the supplied size. -uc nnn Set number of colors in output palette (GIF/PNG) 1-256 (Default=256) -ud nnn Set % dithering (GIF/PNG), 0-100 (Default=0) -x[f] Keep Extra color. By default all color images are written with downsampled color. This option lets you keep extra color information (at the expense of file size). If you add an f (i.e. -xf) all JPEGs written will be made extra color. This might be useful in making high quality thumbnails. -g[f] Keep Progressive. Use this switch to preserve progressive JPEGs. If you don't specify this switch all JPEGs will be written as standard non progressive JPEGs. If you add an f (i.e. -gf) all JPEGs written will be made progressive. Progressive JPEGs are sometimes a little smaller but the progressive effect is not always desirable. Status Options -s[o][v][f][s][l] [filename] -so Don't write to standard output. Useful if you are running from another program. -sv Verbose progress. Write out the file name as it is compressed. -sf Display current file number on screen. Useful for monitoring progress with a lot of files. -ss Write Summary Statistics at the end i.e. Storage and estimated cost saving. The cost savings are based on web traffic costs at $300 for 1G a day and 100 downloads a day of each image. -sl filename Write output to a log file. This will instruct JPEG Disk Optimizer to write it's output messages to a file. This is useful if you are running it from another program or want a log of any images that failed to be compressed. Note fatal errors are sent to the screen. Output Options -o[j|t][r][x][a Text] Output file options -oj Output JPEGs (default) -og Output GIFs -on Output PNGs -ot Output TIFFs (24 bit) -os Output Output same type as read in (JPG>JPG, GIF>GIF etc.) -or Overwrite read only files -ox Don't create the output images, e.g. for reporting potential savings -oa text Append "text" to each file name eg -oa _T = file.jpg => file_T.jpg -od xx yy Set X and Y resolution to xx and yy DPI (default - same as input) Clean Image Options -cn nn Remove Noise and spots from image before compressing. -cn 0 means de-spot only -cn nn means remove noise where nn is the % amount to remove (1-100) -ca aa r.r ss Advanced remove noise, aa amount in %, r.r radius in pixels, ss Sensitivity in %. Note that you can use the -ca option to perform very powerful image enhancements such as removing compression artefacts, removing dithering, removing background noise prior to watermarking. Noise reduction can also be very effective at helping JPEG compression, you can achieve an additional 10% compression or more and often improve the image. It is best to experiment with different values but you can try -ca 50 2 20 Resize and Thumbnailing Options -rh[m] nnnn Resize the image to have height nnnn pixels. -rw[m] nnnn Resize the image to have width nnnn pixels (note if only height or width is specified the other dimension will be scaled proportionally to maintain the images aspect ratio) If option "m" is used no enlargement will be carried out -rb[#rrggbb] Resize the image to fit in a box whose size is given by the -rh and -rw switches. Maintain the aspect ratio of the image. If a color argument is supplied (in hexadecimal HTML format) the image will be padded out with this color to the size specified by the -rh and -rw switches. If only height or width is specified the other dimension will be set to the same as the other. -rk n.n Resize sharpness. When resizing images it is usually best to sharpen them a little. Use this switch to override the default setting of 0.3. 0.0 is no sharpening, 1.0 is maximum sharpening. -rc xo yo ww hh Crop image with xoffset yoffset width height Visible Watermarking Options -vf filename Visible watermark file name (.PNG, .GIF, .TIF or .JPG) Specifies a smaller image to place on the image (Note alpha channel transparency is supported) -vp Specify that TIF is of type with pre-multiplied alpha (Normal case for paint packages). Note: pre-multiplied alpha TIF files are easier to create but do not give quite as good results -vo nn Visible watermark opacity in % (default 100%) -vl pp Visible watermark location, (default top left) here pp is: +--------+--------+--------+ |0 or tl |1 or tm |2 or tr | +--------+--------+--------+ |3 or ml |4 or m |5 or mr | +--------+--------+--------+ |6 or bl |7 or bm |8 or br | +--------+--------+--------+ eg -tl 5 or -tl mr for middle right -vm nnn nnn Horizontal and vertical margins in pixels (default 20) -vt#rrggbb Visible watermark transparent color (rgb hex) If specified all pixels in the visible watermark file that match this color will be treated as transparent. -vs Snap to JPEG 8x8 or 16x16 block (improves edges) Enhancement -ll nnn Set % automatic levels adjustment (Default 0=OFF) -lc nnn Set % automatic color adjustment (Default 0=OFF) -ls nnn Set % sharpen (Default 0=OFF) -lb nnn Set % brightness adjustment (Default 0=OFF) -ln nnn Set % contrast adjustment (Default 0=OFF) -lg nnn Set % gamma adjustment (Default 0=OFF) Input Image Information Reporting Options -i[g][w][c] Output Image Information These switches control how jdo outputs information. -ig Output general image Info e.g. width, height, filesize -iw Output watermarking Info -ic create machine read CSV data with header rather than human readable Preferences Options -pj Only compress JPEGs (skip TIFFs). This will force a compatible behaviour with earlier versions of jdo. -a switch: advanced options Use this file to get more control over how images are compressed. This is particularly useful if you are compressing JPEGs from a variety of sources at a variety of compression ratios. For example you will probably want to instruct JDO not to re- compress images that are already highly compressed. Here is an example file: # keep progressive if input file has it 1 1 0 x p # no change 50 50 0 x p # no change, no MagiCompress 70 70 30 x p # A little MagiCompression 80 80 50 p # medium MagiCompress 90 90 60 p # no extra color, 60% MagiCompress <6000 # Skip files smaller than 6000 bytes %2 # Skip file if jdo can't compress it by more than 2% The first line is a comment. Anything after a # character will be ignored. The following five lines are part of a look up table to tell JDO how to compress images. You can put up to 100 entries in the table corresponding to the 100 possible values of quality (1-100) but if you leave entries out jdo will fill in the gaps for you. The 3 numbers across are: Input quality, Output quality and MagiCompression. All three are required. Optionally you can follow this with an x and/or a p which will tell JDO to keep JPEGs with Extra color and progressive JPEGs. In the above example we have put a p in each line so that all JPEGs that are read in as progressive will also be output as progressive. In the above example from input quality 1-50 the output quality will be the same as the input quality and no MagiCompression will be used. This together with the x flag stops JPEGs in this range being compressed further. Between input quality 50 and 90 the output quality is kept the same as the input quality but MagiCompression is gradually increased. Any extra color information removed after quality 75. Between input quality 90 and 100 the output quality is reduced to 90 and 60% MagiCompression is used. See Appendix A. to see how JDO expands the above table into a full 100 entry table. The next entry: <6000 # Skip files smaller than 6000 bytes Instructs JDO to skip files smaller than a certain size. This is useful if you only want to compress large graphics. The next entry: %2 # Skip file if jdo can't compress it by more than 2% Sets the size at which JDO skips the file if it is unable to compress the file significantly (for example it may already have been compressed with Image Optimizer). In the above example images are skipped if jdo can not compress them by at least 2%. Note the default value is 3%. Sample applications Compress all new files overnight If you are running a free hosting service you might want to run an overnight job to compress your users images automatically. This can save you a very large amount in bandwidth charges. It is important not to compress the images more than once so you should use the -n flag. A typical command line would be: nice find /tmp/images -type f -iname '*.jpg' -exec jdo -n -q 90 -m 40 {} \; On the first time of compression jdo puts an (invisible) marker "JDO" in the file. The next time jdo is run it will detect the marker and skip the file. Notes: Compresses all jpg files in /tmp/images and below "-type f" ensures that only files match, directory names with a jpg extension will not be passed to the exec section of the command. "-iname" causes files with capitalised extensions to match. In this application we recommend the use of a config file to avoid over- compressing already highly compressed JPEGs (see above). Purchasing JPEG Disk Optimizer You have been granted without charge the right to use the evaluation version of JPEG Disk Optimizer (jdo.exe) for the period of 30 days for evaluation purposes only. You may not distribute the compressed images until you have obtained a licence from xat.com for the commercial product. Use of this software after the 30 day period is strictly prohibited. If you wish to purchase JPEG Disk Optimizer contact xat.com for details - jdoinfo@xat.com. When you have purchased JPEG Disk Optimizer you will receive the full product by email. Two versions are available, with and without MagiCompression and noise reduction. The reduced features version does not support the -m or -c switches. Note: The evaluation version also adds a small "xat.com" to the top left corner of the image. License Agreement: JPEG Disk Optimizer Copyright © 1999-2006 by xat.com ltd. (www.xat.com) All rights reserved. Your use of this software indicates your acceptance of this license agreement and warranty. Commercial Version One commercial copy of JPEG Disk Optimizer should be licensed per system. Governing Law This agreement shall be governed by the laws of England. Disclaimer of Warranty The authors make no warranty or representation, either express or implied, with respect to this software, its quality, accuracy, merchantability, or fitness for a particular purpose. This software is provided "as is", and you, its user, assume the entire risk as to its quality and accuracy. This software is sold without warranties as to performance of merchantability or any other warranties whether expressed or implied. No warranty of fitness for a particular purpose is offered. Any liability of the seller will be limited exclusively to product replacement or refund of purchase price. Distribution The Evaluation Version may be distributed unmodified provided any charge is to cover distribution costs only. The Evaluation Version may be placed on CDs or magazine floppy disks etc. as long as xat.com ltd. (info@xat.com) are informed. Appendix A How the config file table is expanded The following table: 1 1 0 x p # no change 50 50 0 x p # no change, no MagiCompress 70 70 30 x p # A little MagiCompression 80 80 50 p # medium MagiCompress 90 90 60 p # no extra color, 60% MagiCompress is expaned to this by jdo internally: 1 1 0 x p 2 2 0 x p 3 3 0 x p 4 4 0 x p 5 5 0 x p 6 6 0 x p 7 7 0 x p 8 8 0 x p 9 9 0 x p 10 10 0 x p 11 11 0 x p 12 12 0 x p 13 13 0 x p 14 14 0 x p 15 15 0 x p 16 16 0 x p 17 17 0 x p 18 18 0 x p 19 19 0 x p 20 20 0 x p 21 21 0 x p 22 22 0 x p 23 23 0 x p 24 24 0 x p 25 25 0 x p 26 26 0 x p 27 27 0 x p 28 28 0 x p 29 29 0 x p 30 30 0 x p 31 31 0 x p 32 32 0 x p 33 33 0 x p 34 34 0 x p 35 35 0 x p 36 36 0 x p 37 37 0 x p 38 38 0 x p 39 39 0 x p 40 40 0 x p 41 41 0 x p 42 42 0 x p 43 43 0 x p 44 44 0 x p 45 45 0 x p 46 46 0 x p 47 47 0 x p 48 48 0 x p 49 49 0 x p 50 50 0 x p 51 51 1 x p 52 52 3 x p 53 53 4 x p 54 54 6 x p 55 55 7 x p 56 56 9 x p 57 57 10 x p 58 58 12 x p 59 59 13 x p 60 60 15 x p 61 61 16 x p 62 62 18 x p 63 63 19 x p 64 64 21 x p 65 65 22 x p 66 66 24 x p 67 67 25 x p 68 68 27 x p 69 69 28 x p 70 70 30 x p 71 71 32 x p 72 72 34 x p 73 73 36 x p 74 74 38 x p 75 75 40 x p 76 76 42 p 77 77 44 p 78 78 46 p 79 79 48 p 80 80 50 p 81 81 51 p 82 82 52 p 83 83 53 p 84 84 54 p 85 85 55 p 86 86 56 p 87 87 57 p 88 88 58 p 89 89 59 p 90 90 60 p 91 90 60 p 92 90 60 p 93 90 60 p 94 90 60 p 95 90 60 p 96 90 60 p 97 90 60 p 98 90 60 p 99 90 60 p 100 90 60 p Appendix B Introduction to Digimarc® watermarking What is a Digimarc-enhanced image? It's an image that contains a hidden message. That embedded message gives the image powerful new roles in the areas of tracking, Internet marketing, asset management and copyright communication. The hidden message in a Digimarc- enhanced image is imperceptible under normal use, yet easily detectable by software and other devices. The message contains a small amount of digital information, carrying attributes and a set of unique IDs. Why you should use Digimarc-enhanced images Anyone who licenses or publishes images, or uses them in a marketing campaign, can benefit from Digimarc-enhanced images. A national museum currently uses them to communicate ownership of its often-copied artwork. A record company tracks the use of its musicians' photographs throughout the Internet. A major stock photo agency embeds its company name and contact information into its digital photos. These Digimarc- enhanced images serve as "mini-ads" for the company. What they can do At the most basic level, a Digimarc-enhanced image is "marked" as uniquely yours. The embedded message tells the world that you own the image's copyright. But Digimarc- enhanced images are also "smart images." Their special embedded message actually allows them to do things. Digimarc-enhanced images can: ? Help you find your images, wherever they go: Digimarc-enhanced images become homing beacons on the World Wide Web. They allow you to track where and when your images have been moved or copied. ? Take you places: More than just a hyperlink, Digimarc-enhanced images can take viewers to a web site, a licensing data base, an image archive, an internal asset management system, and much more. ? Instruct software and hardware: Digimarc-enhanced images can invoke copy protection (in video, for example) and other automatic responses. ? Announce your copyright, as previously described. ? In addition to all that, despite their embedded messages, Digimarc-enhanced images look untouched. This is because the hidden message is buried subtly in the image's pixels. At very high magnification, the message appears as a slight change in the image's grain or texture. ? ? How to start using Digimarc-enhanced images ? ? First visit http://www.digimarc.com/xat.com with Digimarc to receive your unique Digimarc ID. This ID links your images to a contact profile in Digimarc's database or to an Internet destination. You then embed Digimarc messages with your ID into your images, transforming them into Digimarc-enhanced images. ? From there, you publish your Digimarc-enhanced images just like regular images. Viewers who see your images can use a widely available Digimarc reader to detect and interact with them. The good news is that millions of people can already interact with a Digimarc-enhanced image, since a Digimarc reader is available at no charge on the Digimarc web site. ? It's even possible to interact with Digimarc-enhanced images directly through a Web browser. Using Digimarc Mariner, users can turn images into visual bookmarks and explore the smart features of Digimarc-enhanced images. No matter where the images are published, a single click can take the viewer of your Digimarc-enhanced images to information related to them. Digimarc can also help you track the use of your Digimarc- enhanced images on the web with the MarcSpider service. ? ? How Digimarc-enhanced images work ? ? Digimarc-enhanced images are "smart" because of what's inside them. Each Digimarc- enhanced image carries a digital message that is imperceptible under normal use, but detectable by computers and software. Digimarc's technology embeds the message by making subtle changes to the brightness of the pixels in an image. Because the embedded message is carried by the image's pixels, it is independent of the file's format. It can survive normal image edits, file format transformations, copying, printing and scanning. ? ? Inside information ? ? The message embedded in a Digimarc-enhanced image carries identifying information, attributes or instructions. In Digimarc's Internet imaging system (used for copyright communication, asset management and marketing & e-commerce), a unique Digimarc ID; an image ID, transaction ID or copyright year; and a set of image attributes is embedded. This information links to places and information, uniquely identifies the image, and communicates information about the image to applications. ? ? Repeated for robustness ? ? One of the patented features of a Digimarc-embedded message is that the pattern containing the message is repeated over and over throughout the entire image. This repetition strengthens the message so it can be recovered even after the image is cropped or otherwise changed. ? While pattern repetition provides part of the durability solution, Digimarc-enhanced images must also survive more radical transformations such as scaling and rotation. To address this need, Digimarc created a patented technique that adds a hidden "orientation pattern" to the image. Using this special feature, Digimarc's reader can recover the watermark after scale changes of as much as .6 times to 2 times the original image size. A Digimarc watermark can also survive any amount of image rotation. ? These features are important for watermarks to be able to survive changes from a digital to an analog format and back again. Examples include outputting a digital image to print, then scanning it back into digital form; or recording a digital video to VHS format and then back to digital form. ? ? Striking a balance ? ? All of this work necessitates an important balance. Clearly the increased functionality provided by the embedded message is of great value, but the image itself must always be the main consideration. While the embedded message must be sufficiently robust to withstand the expected image modifications, it must also remain imperceptible under normal use. That's why another of Digimarc's patented innovations is the ability to make Digimarc messages "perceptually adaptive." They conform to the unique pattern of each image, placing more energy in areas with rich detail and less energy in flat areas. This process ensures that Digimarc messages remain durable without being easily visible. ? ? Notes on durability, quality and compression ? ? While JPEG Disk Optimizer automates the embedding process as much as possible for you, you will still need to make decisions concerning message intensity, durability, image quality and compression level. It's important to remember that each of these elements affects the others. ? When working with your images, you can instruct JPEG Disk Optimizer(with the -wm switch) to determine whether or not the message has been successfully embedded by measuring its durability, which can be thought of as the ability of the message to survive image modifications such as cropping, scaling, rotating, etc. If JPEG Disk Optimizer finds that an embedded message has an unacceptable durability, it embeds the message again at a higher intensity. Intensity is the amount of energy that is applied in embedding the message; in other words, the degree to which the image's pixels are modified in order to encode them with the message. ? An increase in intensity, then, increases the durability, but it may also decrease image quality slightly by making the pixel changes more visible. When the changes become visible, they can appear as a slight texture or grain in the image. Similarly, a decrease in intensity results in better image quality, but the embedded message is less durable. Because of this interaction between durability and image quality, it is the second priority of JPEG Disk Optimizer to preserve image quality as much as possible. ? When you add JPEG compression to your embedding operation, the balancing act between message durability and image quality becomes a bit more challenging. Since JPEG is a lossy compression method, it has an obvious negative effect on image quality. It can also affect message durability, since any degradation of the image itself will also degrade the robustness of the embedded message. Of course, the impact of these variations is dependent on your chosen compression level and the actual content of the image. In turn, the level of compression you choose is usually influenced by how the image will be used. For example, images that will be posted on a Web site are usually compressed rather aggressively to speed up their display in the user's browser; as a result, it may be quite acceptable to embed at a higher intensity with these images - the resulting durability helps to ensure the persistence of the embedded message, and the increased visibility will often not be noticeable with medium-resolution JPEG images. ? JPEG Disk Optimizer is designed to help you balance these factors, providing you with controls that affect this important interaction. ? Use the -we switch to specify the starting intensity ? Use the -wm switch to specify the desired durability ? Use the -ws switch to specify the intended DPI JPEG Disk Optimizer will then work iteratively to achieve the desired durability by: 1.Increasing the intensity 2.Decreasing the MagiCompression 3.Increasing the Quality Appendix C Digimarc® watermarking FAQ This section contains answers to common questions you may have about understanding Digimarc watermarking. Subscriptions/Digimarc Ids Q. How do I get a Digimarc ID? A. You can register for a Basic or Smart Digimarc ID in one of two ways: 1) Go to the Subscription and Digimarc ID (http://www.digimarc.com/xat) section of Digimarc's web site; you will find a link to the subscription forms there. 2) Have a registration form faxed to you by calling +1-503-495-4623 Q. Why does JPEG Disk Optimizer not support Distributor IDs? A. The Distributor ID is no longer supported by Digimarc. Q. Where is the contact information and smart image link information stored? A. Contact profiles and smart image link information are stored in Digimarc's database. The system is accessible via a high-speed Internet connection, and is operated in a data center with full 24-hour operations and support. Q. How do Smart IDs work? A. First, you embed your Smart ID in your images using JPEG Disk Optimizer, and maintain a destination link in Digimarc's database (which tells the image where to send people). Then, viewers of your images can navigate to the smart image destination by using a Digimarc reader. This is accomplished behind the scenes by first going to Digimarc's database, and then redirecting to your chosen URL. This also means that you can change the destination URL at any time, without changing the embedded message in the image. Q. Who's responsible for maintaining the records? How often can I change my Smart ID links? A. Digimarc customers can update their contact profiles and Smart ID links at any time. Simply go to Customer Service (https://www.digimarc.com/db/owa/member.startup) in the Service & Support area of our web site. Once there, choose the "update profile information" option and enter your Digimarc ID and password to log in to your account. Q. Who controls the data someone sees about me? A. At the time you provide us with your contact details, you control the information you want viewers to see. You can exclude your address, phone number, and other specific information. You're in charge. Q. What happens to my Digimarc ID if I decide not to renew my subscription? A. Your Basic ID is yours forever (subject to the terms and conditions of the subscription agreement). However, you are responsible for maintaining current contact information. Your Smart ID will remain active as long as you maintain your subscription. If you let your subscription lapse, it will be automatically converted to a Basic ID. Q. I lost my PIN or password. What can I do? A. To retrieve your PIN, simply go to Customer Service (https://www.digimarc.com/db/owa/member.startup) in the Service & Support area of our web site. Once there, choose the "review account status" option and enter your Digimarc ID and password to log in to your account. If you've forgotten your password or Digimarc ID, you can call Digimarc customer service at 800-DIGIMARC (800-344-4627), or +1-503-495-4623 from outside the US. Q. If I use more than one Digimarc embedding tool, do I need to get a different Digimarc ID for each application? A. No. You can use your Digimarc ID with any Digimarc embedder. Some customers choose to use multiple IDs to associate images with different brands, or to provide different Smart ID destinations. General Technical Questions Q. What if I Digimarc-enhance all my images and then Digimarc changes its technology? A. The embedded message carries version and message type information, and Digimarc products have been designed to support upward compatibility. Also, Digimarc readers are able to recognize previous message types. If a Digimarc reader finds a newer version of a Digimarc embedded message that it does not support, you will be notified and instructed to download the most recent reader. Q. Do embedded Digimarc messages survive printing? A. Yes. This was a primary design goal of the technology and has been successfully tested. Once a Digimarc image has been printed, it can be read again by scanning the image on a digital image scanner. For best results, we recommend using Digimarc utilities as part of the prepress process, once the image has been prepped for printing. Q. What image file formats can be Digimarc-enhanced? A. You can use all the formats supported by JPEG Disk Optimizer i.e. JPEG and TIF images. You can use JPEG Disk Optimizer to check that the watermark has survived compression, color reduction and dithering. Q. What color spaces are supported? A. The Digimarc embedding process is color space independent. You can embed a message into RGB and grayscale images using JPEG Disk Optimizer. Q. How small an image can I Digimarc-enhance? How large? A. If you don't intend to apply any compression or other image edits, the image can be as small as 100 x 100 pixels. However, if you believe your image might be compressed, cropped, rotated, scaled or otherwise modified, we recommend a minimum size of 256 x 256 pixels. There is no upper limit on the size of an image that can be Digimarc- enhanced. Q. Do Digimarc-enhanced images survive compression? A. This was a design goal of the technology. Lossy compression methods such as JPEG actually remove some image data in order to decrease file size. This can have varying effects on survival of the embedded information. Generally if you favor image quality over image size when selecting the degree of compression, you'll get the most favorable results. You may find you have to use less than normal MagiCompression for the watermark to survive. Q. Can I Digimarc-enhance a JPEG image? A. Yes, however, in general, you'll get the best results and quality by starting with an original, uncompressed image. Q. Can the embedded Digimarc message be removed? What happens to the image if it is removed? A. Digimarc does not provide tools or software to remove its Digimarc messages from images. The message embedded in a Digimarc-enhanced image is robust, so it survives normal image edits, file format transformations, and copying. However, it isn't intended to be absolutely permanent. A skilled person could remove it from your image if he or she really worked hard to do so. When your Digimarc-enhanced image is opened in an editing tool, it reminds the viewer that it is copyrighted. As a result, the act of intentionally removing your Digimarc message would be an obvious attempt to infringe your copyright, triggering opportunities for stiffer penalties. Q. Can I remove my own Digimarc message from an image? A. No. Since you are the image owner, it is assumed that you will retain an original, non Digimarc-enhanced image in the event it is needed. Q. What happens if I try to add a new Digimarc message to an already Digimarc- enhanced image? A. You will be prevented from adding another watermark if one is found in the image. MarcSpider Q. How do I access my MarcSpider report? A. MarcSpider reports are available online any time of day. Simply go to Customer Service (https://www.digimarc.com/db/owa/member.startup) in the Service & Support area of our web site. Once there, choose the "view MarcSpider report" option and enter your Digimarc ID and password to log in to your report. Q. How often is my MarcSpider report updated? A. Reports are updated continually, so you'll always see the latest image finds. You can view your as often as you like, with options to view finds for the current month or by date range. Acknowledgments Digimarc is a registered trademark of Digimarc Corporation This software is based in part on the work of the Independent JPEG Group. This software is based in part on the Image Magick library. Portions Copyright (c) E. I. du Pont de Nemours and Company The software is provided "as is", without warranty of any kind, express or implied, including but not limited to the warranties of merchantability, fitness for a particular purpose and noninfringement. In no event shall E. I. du Pont de Nemours and Company be liable for any claim, damages or other liability, whether in an action of contract, tort or otherwise, arising from, out of or in connection with ImageMagick or the use or other dealings in ImageMagick. This software includes Sam Leffler's TIFF Software: Portions Copyright (c) 1988-1996 Sam Leffler Portions Copyright (c) 1991-1996 Silicon Graphics, Inc. THE SOFTWARE IS PROVIDED "AS-IS" AND WITHOUT WARRANTY OF ANY KIND, EXPRESS, IMPLIED OR OTHERWISE, INCLUDING WITHOUT LIMITATION, ANY WARRANTY OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. IN NO EVENT SHALL SAM LEFFLER OR SILICON GRAPHICS BE LIABLE FOR ANY SPECIAL, INCIDENTAL, INDIRECT OR CONSEQUENTIAL DAMAGES OF ANY KIND, OR ANY DAMAGES WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER OR NOT ADVISED OF THE POSSIBILITY OF DAMAGE, AND ON ANY THEORY OF LIABILITY, ARISING OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.