Digital Video for Science Erik Meesters The Netherlands Institute of Sea Research P.O. Box 59, 1790 AB Den Burg The Netherlands Ph. 31-(0)-222-369530 Fax 31-(0)222-319674 Email erikm@nioz.nl Table of Contents (to quickly go to the page you want, double-click on the page number) 1. INTRODUCTION 2. FROM LIGHT TO DIG 3. DIG PHOTO CAMERAS VERSUS DVS 4. TRANSFER OF DIG DATA 5. IMAGE ANALYSIS 6. NOTES FROM USERS 6.1 EXTRA UW-LIGHTS OR NOT? 6.2 HIGHEST QUALITY IMAGE 6.3 INSTRUCTION TAPE FOR DIGS 6.4 FAILING AUTO-FOCUS 6.5 DCR-VX700 VS. DCR-VX1000 (ABOUT U$ 2000 DIFFERENCE) 6.6 CUSTOM MADE UW-HOUSINGS 6.7 PHOTO CAMERA 6.8 CHEAP DEVICES FOR UW 6.9 AMPHIBICO HOUSINGS...... 6.10 DUBBING 6.11 JAPANESE, EUROPEAN, AMERICAN MODELS 6.12 IMAGE GRABBING 6.13 DAMAGED VIEWFINDER 7. OTHER DIGITIZING POSSIBILITIES 8. ACKNOWLEDGMENTS 9. WWW PAGES Before you read further Because I've become allergic to the words digital and video, I'm using abbreviations in this overview. Dig means digital; DV, dig video camera; Photo camera, dig photo camera; cameras are always dig, whether DV or photo cameras, unless otherwise specified. I do not claim to be a specialist on digital cameras, frame grabbers, or image analysis, so don't blame me for any mistakes (including the language). This is an extremely fast developing field, so reported prices are probably already lower. If you profited in any way from this review, you're allowed to send/email a nice postcard to my address. I'd appreciate constructive comments and tips at my email address: erikm@nioz.nl. Introduction Why this review? Scientists often collect images and need to do measurements on these images. Collecting images with dig cameras appears to offer a number of attractive advantages for many research fields. Since most data processing and analysis is dig, direct collection of dig data can offer a great gain in speed of data collection, processing, and data transfer. Dig pictures offer high quality (3 times as much color bandwidth as analogue recording formats) and can be copied instantly without loss of quality. Images area stable in that, unlike film dyes, dig images never fade or change color. Dig photography eliminates the specialist consumables required at each stage of the photographic process - for example, there is no need to store film in refrigerators (more space for beer), no need for bikes, to take the film to and from the lab (no loss of calories) and no more scanning for reproduction (more money for beer). Instead of developing a roll of film, making prints and then scanning them, dig pictures can be transferred to computers directly without loss of quality. Images are available for viewing instantly on screen and ready for import into a project. Duplicate images are identical to the original in every respect and can be made, on demand, without any special equipment or material. This, in turn, cuts down the administration and paperwork that has to be processed by office and accounts staff, if you have any (otherwise, you might be able to use them more efficiently or, alternatively, fire them). Without a doubt, all of us will be using dig image capturing devices in less than 5 years, if not for data collection than for pictures of our vacations. There are already DVs that are smaller than your normal photo camera. So if you need to buy a new DV, dig is the way to go. Before buying anything you should determine exactly why you need the camera: what data do you want to collect and how are you going to analyze your data. A very important question is whether you will need to analyze moving images or just still pictures. From light to dig Dig cameras are similar to their film-based brethren, except that incoming light hits a charge-coupled device (CCD) and the resulting electrical signal is stored digly. CCDs are light-sensitive integrated circuits with a mosaic of thousands or millions of pixels. Place one in a camera's image plane, where the film would normally go, and the light falling onto it causes a minute electrical charge to be held in each pixel of the CCD. These charges are analogue, because they are proportional to the brightness of the light. When read in an ordered way and amplified, the pixels can be displayed in the same order, reconstructing the original image. CCDs elements are arranged in a variety of ways. Linear arrays consist of light-sensing elements arranged in a single row. Tri-linear arrays are three such rows, each with a red, green or blue filter in front of it, usually cemented together. Area arrays have their separate light-sensing elements arranged in a two dimensional grid pattern. They are suitable for capturing monochrome images instantly. With suitable filtration they can be used to analyze color. Dig photo cameras versus DVs Because dig handycams are offering greater flexibility, sufficient image quality, and lower overall price, dig camcorders may soon eclipse dig still cameras as multimedia production tools and surely as digital image capturing devices. The transfer from stored image to computer appears to be better designed in photo cameras. Some DVs still have no dig output, while the pc-cards from the photo cameras take your digital image directly to the computer. If no dig output is available, the DV signal needs to be transferred via the analogue output. The device at the other end of the cable then converts the analogue pictures back into dig for editing. Each analogue/dig conversion deteriorates image quality, however, a point to remember is that the analogue output from a DV is still of superior quality when compared with analogue brothers. To the unaided eye the difference in quality between the dig image and the direct analogue output from a DV is indiscernible. Also, many frame-grabbers have no dig input, so if you buy a DV and you want to get the most out of it, make sure you use a dig frame-grabber as well (e.g. DV-Master from Fast). Sony DVs have a dig output and Sony offers a framegrabber that works with the DV output. Sony has also just released its dig VCRs. The DV units are as portable as any of the traditional camcorders on the market, and they delivery a very good image quality -- resolutions as high as 500 lines, almost twice that of standard TV. Some models even have a still-image mode that generates higher resolution. Like dig still cameras, images are available for immediate preview and use. And, obviously, dig camcorders can capture full motion as well as stills. Three-CCD models from Sony and Panasonic are priced at less than $4,000. They include such features as 10X optical zoom and 20X dig zoom, shutter-speed control and color viewfinders. Less expensive models, each using only one CCD chip, sell for around $1,000. Dig. photo camera. cameras generally have a higher resolution (more pixels), but their price is much higher, especially for the higher resolution cameras, and the storage capacity of dig. photo camera. cameras is a lot less than that of DVs. DVs often have a 10/20x zoom so you can improve image quality by making close ups. DVs use tape, photo cameras use mostly (more expensive) pc-cards. Storage capacity on tapes is much higher than on pc-cards. Another problem with photo cameras is the limited availability of underwater housings. If analysis is to be done on the computer screen, image quality of DVs is probably sufficient in most instances. Who needs to analyze wall-size pictures? Because I feel that at the moment most people will be more interested in dig video cameras than in dig photo cameras, the rest of this overview will only discuss dig. video cameras. Transfer of dig data To transport dig data a new standard has been developed, however, most vids do not yet support this standard. Since 1,393 standards had already been considered, the new standard became known as IEEE 1394. Apple preferred a less numeric name and called their version of IEEE 1394, FireWire (tm of Apple Computer). For more information consider the WWW addresses given at the end of this overview. There are only a few plug-in cards that work with the DV interface. Most cards still need the analogue input that all vids offer. Sony has a card that uses the DV signal from its DVs, the DVBK-1000, but you can only use it to grab single images. The PCI board DV MasterTM from Fast accepts dig data directly from digital (DV) camcorders and VCRs in real-time. Sony has also developed a Digital VCR, which has just come out. For more details you should take a look at the web pages given below. Image analysis There are a number of image analysis programs that may be used together with your vid. One of the most popular (public domain) programs is NIH Image (developed at the U.S. National Institutes of Health and available on the Internet at http://rsb.info.nih.gov/nih-image/). Image directly supports four frame grabber cards for acquiring images from video sources, such as TV cameras and VCR's: the Data Translation QuickCapture, the Scion LG-3, Scion AG-5, and the Scion VG-5. All allow brightness, contrast and pseudocolor to be adjusted during "live" capture. All allow continuous histogram display and "live" paste, which can be useful for image alignment. Notes from users Extra UW-lights or not? Apparently, the high-end DVs function well under low light intensities. Under clear conditions a blue light filter renders the images in 'true' colors down to 20m. This may be especially useful under turbid conditions were back-scatter of vid lights may render your images useless. Highest quality image If you only want the highest quality, you can have your pictures transferred on a CD, which will give you the highest resolution. This is expensive, and is only done by dealers. Instruction tape for digs I would suggest you purchase a VHS instructional vid tape on this camera. It's a 100 minute program about the camera and how to use it (the SONY manual is difficult to read). The tape title and the company that sells it is: SONY's VX-700 and VX-1000 Dig Vid Camcorders $49.95 +SH item: DV-CAM-2 Knowledge Path vid 1630 N. Main street, Suite 132 Walnut Creek, CA, 94596 1-800-537-3050 or 1-510-932-2282 e-mail: knowledgepath@msn.com Failing auto-focus One problem I have encountered taping underwater occurs in shallow water when there is "flickering" light. If you set up the camera to run on a tripod under these conditions, the stationary objects on the edges of the frame appear to wave slightly. It is a very annoying effect and I think it is caused by the auto iris compensating for the sudden increases and decreases in light and thus changing the depth of focus of the lens (or something like that). The way to solve it, I think, is to over-ride the auto-exposure control and use manual exposure. Another way, apparently, is to put your camera on the maximum depth of field position. DCR-VX700 vs. DCR-VX1000 (about U$ 2000 difference) My organization has just purchased a Sony DCR-VX700 DV with a Stingray VX700 housing and Sunray 5500 lights to make underwater counts of fish populations. We considered the VX1000 but were advised that the difference in quality between the two was extremely difficult to detect underwater. A 3 CCD camera may have been marginally better, but we felt we couldn't justify the extra cost. Custom made UW-housings We have built underwater housings for COHU CCD high resolution vids using microscope optics to photograph micro bubbles in breaking waves. If you need information about uw connectors or mounting hardware we can help you, also we use a software package that is very powerful in regards to real time analysis of vid data. Photo camera The developments are faster than I can type. At the moment there's a photo camera by Kodak ( DC120) with 1.2 million pixels, costing only about U$ 1200 (comparable to the polaroid that cost U$ 5000 only 6 months ago). But again, no underwater ready-made housing. Cheap devices for UW EWA underwater bag which I strongly DO NOT recommend--in about 7 months I've been through four bags--they keep leaking on me. The bags are designed for recreational use and do not withstand daily use. The camera is great with fantastic resolution and works great in low light conditions. Amphibico housings...... Although most people are happy with these houses, people that use the Amphibico housing intensively have had problems with it. All houses seem to have their problems every now and then, however, it's good to know what to expect if you're out their under water. They use to be a company with a very strong product and equally strong reputation... it would seem that with this latest housing they have some problems.... and are not prepared to face up to it if reports I hear are to be believed. The main problem on the housing it seems are the small green buttons that operate the camera functions. I know of several instances where these buttons have "popped" off and flooded the housing... destroying the camera inside. This happened to me recently on a filming trip to the Red in March. I was using multiple cameras for the shoot using three VX1000 and Amphibico housings. The button on the back of the housing "popped" off whilst I was operating it. Fortunately I still had my finger on the button and, realizing that if I let go of the button water would flood into the housing I managed to surface.... a small amount of water inside and luckily no damage. Two of my colleagues were not so fortunate and destroyed their cameras with the same problem.... one only last week!! Other problems I have encountered... Intermittent problems with electronics controlling the camera functions.. Water flooding into the handles Buttons on the handles breaking off. Of the three camera systems I used in the Red Sea every one had a problem at some point in the shoot! I have not personally written to Amphibico about these design problems but my colleagues did write and were rebuffed by Amphibico who could only suggest it was the operators fault ... not the design. As a professional cameraman who has used this housing I could not give it my recommendation. It nicely finished and well made in many respects but is simply not reliable enough in my experience and the flooding problem with the buttons is of serious concern. The Sony VX1000 is a nice camera for research work and data collection..... I suggest the Stingray housing as a possible alternative to the Amphibico. In using this housing you have to modify the camcorder by removing the handle. I have not used this housing with the VX1000 but the housings have been used extensively with other cameras for a number of years by the Australian Institute of Marine Science for their field data collection without major problems. Also (other user).... Both Richard and I tested our cameras in our separate Amphibico housings and both of us had difficulties. At 180' my housing stopped functioning. The manual aperture button stuck would not release once pushed to the "on" position, neither would the zoom or focus buttons operate. The only function that would work at that depth was the record-on/off button. Function resumed to normal once I returned to depths above 150' a few minutes later. Likewise, Richard's camera failed to function at even deeper depths. According to the person using it (not Richard) the camera took on a "mind of its own" and began to focus and record etc. without hitting the buttons. These problems seem to be related to the housing and not the camera. Dubbing Regarding the "Firewire" dig connection that Bruce mentioned..... with the PAL version of the camera you can only do dig dubbing to a "home" DV recorder as the camcorder does not have a Record Input like the NTSC version from the USA. The camera otherwise functioned superbly and produces excellent images. Moreover with the digital copying capability of the SONY cameras we could make exact copies between the various cameras available to us in Palau (seven SONY 1000's at one time in one place!) using the FireWire (I noted that someone mentioned the Panasonic digital camera has a better viewfinder which might be useful, but as of now, you cannot make digital copies off of the Panasonic camera which I find to be a distinct disadvantage of the Panasonic camera). Japanese, European, American models The Japanese and European models of the Sony are similar but not the same, so check out the model you want to buy. Image grabbing Screen-machine from Fast, a frame-grabber, is better than the Sony card, according to a salesperson who has used both. A disadvantage is that the Screen-machine uses the analogue output of your DV. However, the primary image from the analogue output from a dig vid is a lot better than the image from an analogue camera. This means that the use of image grabbers that require analogue inputs will still give you better quality than images from non-dig cameras. In this respect, analogue image grabbers (e.g. from Fast or Scion Corporation) may be good alternatives for the Sony dig frame grabber. The Sony dig frame grabber does not appear to be able to show you the image at the size of your computer screen, while for example the Screen Machine does (pers com sales person Fast). I haven't compared the two framegrabbers yet, but will next month. Anyone, who has?. Damaged viewfinder One other point for amateur videophiles like me: don't let the sun directly hit the viewfinder on your camera! This can easily happen on a boat when you pass it up to someone else. The sun leaves nasty permanent greenish smudges on the viewfinder screen! Other digitizing possibilities Slides and photos can be scanned with slide scanners that transform your slide to high resolution dig images. A relatively cheap option is the Nikon Coolscan II with a maximum resolution of 2592 x 3888 pixels (see http://www.klt.co.jp/Nikon). Images can also be read and stored on CD (giving the highest resolution possible, according to Kodak employee). However, this requires the images to be sent away, since the necessary equipment is very expensive. Acknowledgments I like to thank the people that have contributed to this overview: Amanda Bourque Bruce Carlson Christian Lott<100342.3407@CompuServe.COM> David Booth Gerard Nieuwland Helge Peter Vogt <106300.475@compuserve.com> John Kirkwood Paul Marshall Stever Gardner Suzanne Dorsey William W Gardiner Sorry, if I forgot someone. The number of people in the marine science community that are using dig. cameras does not appear to be very high yet. Only 11 answered my call for information, but many more asked for information. However, most people do not digly process their images for data analysis. This is likely to change in the near future, when more frame grabbers that use dig input become available. I would also like to thank all the people that have put information on the WWW. Possibly, I've not included every single address that I've used at the end of this summary. WWW pages Firewire/ IEEE 1394 http://firewire.org/ http://www.vxm.com/index.html http://www.skipstone.com/ Sony VX1000 vid http://www.sel.sony.com/SEL/consumer/camcorder/dcr_vx1000.html Digital photo cameras http://wpxx02.toxi.uni-wuerzburg.de/~winstel/walter/nikon.html#digi http://www.nikonusa.com/corpinfo/press/960911ixlens.html http://www.klt.co.jp/ Nikon/http://www.klt.co.jp/Nikon/EID/index.html U/W housings http://www.scubatimes.com/scubatim/vid_1.html Framegrabbers http://www.2fast4u.com/PRESS/DVM111896.HTML http://www.sel.sony.com/SEL/consumer/ dvf.html http://www.scioncorp.com Image analysis http://www.chemie.uni-marburg.de/~becker/image.html http://www.dac.neu.edu/units/artsSci/MSC/ColorImage.html http://www.dac.neu.edu/units/artsSci/MSC/AdvancedImaging92.html http://rsb.info.nih.gov/nih-image/ nih4: http://rsb.info.nih.gov/nih-image/manual/Contents.html http://www.fastmultimedia.com/AltHome.html Digital Background info http://www.2fast4u.com/DEVICENOTES/DVBACKGRND.HTM 10 9