You are here: Home > Archived: 2003 Review of Image Search Engines
2003 Review of Image Search Engines
February 2003
Please note: This page is an earlier version of our current Review of Image Search Engines, which we have archived for reference purposes only. Links on this page are no longer being checked and may no longer work. Please refer to our current Review of Image Search Engines for the latest information.
- Introduction
- General, specialised and meta- search engines
- Collection-based search engines
- Content-based search engines
- Conclusion
1. Introduction
The phrase ‘Image Search Engine’ is most often used of Web-based services that collect and index images from other sites on the Internet. Image searching is sometimes offered by general search engines, like Google or Altavista, but there are also specialised image search engines - services devoted to indexing images or multimedia. In addition, there are meta- search engines, which pass on search requests to more than one search engine and then bring back the results.
Sometimes ‘Image Search Engine’ is also used to refer to collection-based search engines - services that index a single or small number of image collections. Large digital libraries or commercial stock photo collections, like Corbis, typically offer such search engine-like facilities.
All of the categories of image search engines mentioned above are text-based - their indexes are created from words associated with the images. There have also been attempts to create content-based search engines, which ‘index’ visual characteristics of an image, such as its shape and colour. However, these tend to be experimental and are often limited to single image collections.
This review concentrates on general, specialised, and meta- search engines (section 2), but also covers, briefly, collection-based and content-based engines (sections 3 and 4).
2. General, Specialised and Meta- Search Engines
Image search engines are based on existing search engine technology, but they use additional strategies to identify, categorise and rank images.
A search engine’s indexing of images is done automatically, rather than using human indexers, so it must find ways to guess at the image’s content. It might take into account its filename or any accompanying ‘ALT’ picture tags (these are coded into the HTML page). It might look for clues from the image’s context - for example, the words or phrases that are close to the image, or the ‘META’ tags found at the top of the HTML coding. The characteristics of the Web site and its server will also often be taken into account.
Analysis of an image’s text and context can be used to exclude images as well as include them - for example, an image engine will usually consider an image’s context and associated words when it is blocking out adult material.
In reviewing image search engines, it is helpful to use an evaluation criteria and a standard set of search terms. Although the exercise is still subjective, these will offer some basis for comparison.
| Evaluation Criteria In reviewing image search engines, TASI asked the following questions: |
|
| Scope |
|
| Search Options |
|
| Performance |
|
| Presentation |
|
| Support |
|
| Search Terms TASI tested the image search engines using the following terms: |
||
| Category | Terms and phrases | Rationale |
|---|---|---|
| General search terms | Elephant Flower |
General, relatively unambiguous, terms |
| Specific search terms | Danaus plexippus [monarch butterfly] Cleopatra |
Scientific name to test the indexing and historic figure to test scope and relevancy |
| Ambiguous search terms | West Bank Alexander Pope Madonna |
To test the weighting of the indexes and their handling of phrases |
| Specific images | cmy_01.gif (an illustration from the TASI Web site: http://www.tasi.ac.uk/ advice/creating/image.html) newaeflogo.jpg (logo of the American Eagle Foundation: http://www.eagles.org/) sforza-hours-1.jpg (Illustration from the British Library Web site: http://www.bl.uk/collections/ manuscripts.html) |
Searched on unique files to test the comprehensiveness of coverage. We searched on the filenames themselves and on relevant keywords: “American Eagle Foundation logo” and on “Sforza Hours” (with and without the quotation marks) |
TASI’s review was carried out during the first week of February, 2003. For each search engine reviewed we searched on all of the terms and tried to answer all of the questions - the results are presented here in a summary form for convenience. We have also given overall evaluations (poor, disappointing, good etc.). These are subjective judgements, but are based on the more detailed evaluations and on a comparison of all the engines reviewed.
2.1. Major Search Engines offering an Image Search Option
Currently, the main English-language-based search engines offering image searches are AllTheWeb, AltaVista, Google and Lycos. They are all US-based. Yahoo (strictly speaking a directory rather than a search engine) offers a picture search, but this is drawn from the Corbis collection (reviewed in section 3, below). HotBot and MSN Search do not offer true image search features, they simply enable users to limit their search results to ‘only pages with images’.
AllTheWeb Picture Search / FAST
http://www.alltheweb.com/
AllTheWeb was launched in 1999 and is used as a public showcase for the commercial ‘FAST’ search engine. It is known for its wide coverage and good relevancy ranking.
| Criteria | Overall Evaluation | Comments |
|---|---|---|
| Scope | Broad | Very big and up-to-date. Claims more than 2 billion Web sites, but does not specify number of images |
| Search Options | Good | Simple and advanced search options, though not the same sophistication as its general Web page search. Quietly supports Boolean and phrases, but not wildcards. Can limit a search by file type or colour and filter out adult material. Can search within results |
| Performance | Average | Reasonable performance, with quick and generally useful results. No duplication observed, but the occasional dead link and adult image appeared (despite the adult filter) |
| Presentation | Good | Thumbnails first (up to 24 to a page) with links to a larger image and more information: page title, surrounding text, file type, file size and pixel dimensions. From there users can view the full sized image by itself or in its original context. There are clear warnings about copyright |
| Support | Good | Very good online help and ability to request help via a form TASI’s test words |
| TASI’s test words | Adequate | Failed to locate any of the specific files, but did reasonably well at finding results for the other terms. Some irrelevant and poor quality images, especially for ‘Cleopatra’ and ‘Alexander Pope’ |
Altavista Image Search
http://uk.altavista.com/image/
Established in 1995, AltaVista is one of the oldest search engines and was for many years the best. In the late nineties it put all its energy into developing wider portal services and its search facilities suffered. It was overtaken, first by Google and then AllTheWeb, and is now struggling to make a comeback.
| Criteria | Overall Evaluation | Comments |
|---|---|---|
| Scope | Broad | Big and up-to-date index. Does not specify the number of images or Web sites, but probably less than a quarter of AllTheWeb or Google |
| Search Options | Good | Simple and advanced search options. Supports Boolean, phrases and wildcards and basic limitation by format or colour. Can filter out adult material |
| Performance | Average | Reasonable performance with quick and generally good and relevant results. No duplication observed, but several dead links and the occasional adult image appeared (despite the adult filter) |
| Presentation | Average | Thumbnails first (15 on a page) with links directly to the source or to more information (filename, type, size, colour and location URL). The results are accompanied by some advertising and there is no obvious warning about copyright |
| Support | Average | Some online help information and ability to request help via a form |
| TASI’s test words | Good | Found the British Library’s Sforza image but not the other specific files. Produced good and generally relevant results for the other search terms |
Google Image Search
http://images.google.com/
Google is the favourite search engine of search experts and general Web users. Since it went public in 1998, it has concentrated on providing the best search facilities, resisting the temptation to become a portal, the downfall of its rivals. It currently vies with AllTheWeb for the widest coverage and best relevancy.
| Criteria | Overall Evaluation | Comments |
|---|---|---|
| Scope | Broadest | Very big and up-to-date. Claims to search 425 million images and more than 3 billion Web sites |
| Search Options | Very good | Includes simple and a very good advanced search. Has explicit Boolean and phrase searching, but does not support wildcards or truncation. Can limit by size, file type, colour or Web domain, and filter for adult material in a ‘strict’ or ‘moderate’ way. Can search within results |
| Performance | Good | Quick and reasonably good results. No dead links spotted, but there was a little duplication (same picture, but on different sites with different filenames). Although large numbers of results were reported, Google does not actually allow its users to view more than a few hundred images |
| Presentation | Average | Thumbnails first (20 at a time) which link to a split screen with Google on top and the source page below. Limited information provided (filename, extension, pixel dimensions, filesize and location URL) and there is little mention of copyright |
| Support | Good | Good online help available and can contact by email |
| TASI’s test words | Good | Found all of the specific files and reasonably good results for the other terms. Provided a mix of singer and saint for ‘Madonna’ some good maps and photos of Israel for ‘West Bank’ and did especially well at representing ‘Alexander Pope’ |
Lycos Multimedia Search
http://multimedia.lycos.com/
Lycos was launched in 1994. In 1999 it stopped doing its own indexing and began using FAST’s Web index, so in theory it has the same coverage and relevancy as AllTheWeb (above). However, its picture search returns quite different results from AllTheWeb’s. This is because Lycos has entered commercial partnership with Getty (section 3, below) and always presents Getty’s images first in any search results
| Criteria | Overall Evaluation | Comments |
|---|---|---|
| Scope | Potentially broad | Big, since it draws on the same index as AllTheWeb and adds in commercial images from Getty. Compromised because it always presents Getty’s images first and the other ‘external’ images do not seem up-to-date (they contain many dead links) |
| Search Options | Good | Simple search option only, although it seems to respond to Boolean, phrase searching and wildcards. Can filter out adult material. The search results usefully include suggestions for further search phrases |
| Performance | Mixed | Results are quick and the first few pages of images (from Getty) are reasonably good and relevant. Later pages are generally much poorer and frequently contain missing images and dead links |
| Presentation | Poor | There are two classes of images here: the Lycos images (supplied by Getty), with very small thumbnails, larger watermarked versions and simple metadata (title and photographer); and the ‘external’ images with slightly larger thumbnails and different data (dimensions, file size and location URL). The commercial Lycos/Getty results include clear statements of copyright, but there is no such acknowledgement on the external images |
| Support | Poor | Minimal online help available, but can contact by form |
| TASI’s test words | Disappointing | Failed to find any of the specific files, with the other terms leading to mixed results. The quality of the results was not as good as expected, given their commercial source |
2.2. Specialised Image Search Engines
Currently the main specialised image search engines in English are Cobion visoo, Ditto and Picsearch. Sometimes GoGraph (http://www.gograph.com/) is included in this list, but it is not a true search engine (photographers and illustrators submit their work for inclusion) and is very limited in its scope.
Cobion visoo
http://www.visoo.com/
Visoo is owned by a German company Cobion, who produce similar search products for the German language audience.
| Criteria | Overall Evaluation | Comments |
|---|---|---|
| Scope | Unknown | No clear statement of size or currency |
| Search Options | Average | Offers one search box, but this can limited by format, type or colour, or to pictures of people. Can also exclude adult material and banners from the search results. Searches as phrases by default, but can not cope with quotation marks or any other punctuation |
| Performance | Poor | Slow and error-prone. Only displays a maximum of 30 results, no matter how many it finds. A few dead links observed, and some duplication (of pictures, not of files). The results are poor-to-average in terms of quality and relevance |
| Presentation | Poor | Thumbnails first which link to the full image or to the source. Apart from the image, the only information provided is the file size, pixel dimensions and location URL. There is no mention of copyright, but nor is there any obvious advertising or commercialisation |
| Support | Poor | No online help and no easy way of asking for it. |
| TASI’s test words | Average | Visoo didn’t pick up TASI’s specific images on a filename search (since it doesn’t accept punctuation), but it did find the British Library illustration and the American Eagle Foundation logo when relevant keyword were used. Although it recorded high numbers of hits on TASI’s other standard test words the 30 it delivered for each term were generally disappointing |
Ditto.com
http://www.ditto.com/
Ditto is US-based and has been around for a few years, formerly under the name Arriba. In 2002 they were taken to court for copyright infringement and are now much more careful about the way they link to other people’s images. Their index is created through a mix of automatic and human editing.
| Criteria | Overall Evaluation | Comments |
|---|---|---|
| Scope | Unknown | No clear statement of size or currency |
| Search Options | Poor | Only a simple search available, with no filtering or evidence of sophisticated search support. Ditto claims not to include adult images |
| Performance | Average | Slow and small number of hits, but reasonably good and relevant. However, there were numerous dead links and quite a lot of duplication (albeit of pictures rather than files). Despite its family-friendly claim, some adult images cropped up in the results |
| Presentation | Average | Thumbnails first (6 results on the first page and 9 on subsequent) which link directly to the source. Also some commercial images are presented first, though these are visually distinct from the main results. Only information given is filesize - apart from the commercial images, which add a title. There is a copyright disclaimer at the bottom of each page and the thumbnails link to new windows - both consequences of the law suit |
| Support | Average | Minimal online help available, but can contact Ditto via a form |
| TASI’s test words | Average | Ditto failed to find any of the specific files, but did quite well on the others, apart from ‘Dananus plexippus’ and ‘Alexander Pope.’ Instead of the writer, Alexander Pope resulted in a mix of papal, ancient Macedonian, and sports imagery |
Picsearch
http://www.picsearch.com/
Picsearch is owned by a Swedish company. It prides itself on its user friendliness and family friendliness - adult material is automatically excluded.
| Criteria | Overall Evaluation | Comments |
|---|---|---|
| Scope | Unknown | No clear statement of size or currency |
| Search Options | Good | Offers simple and (moderately) advanced searching. Boolean, phrase and truncated searching are supported, but not advertised. Can limit by colour and filesize. Adult material is automatically excluded |
| Performance | Good | A little sluggish in delivering the results, but these were generally very good and relevant. No duplication or dead links were observed, but the occasional adult image appeared in the results despite its efforts to exclude them |
| Presentation | Good | Similar to Google: 15 thumbnails first and then links to a split window with Picsearch on top and the source below. Information includes pixel dimensions, filesize, file type and colour info - including number of colours in the image. There is a clear statement about copyright and no sign of advertising |
| Support | Good | Good online help available and can contact Picsearch via email |
| TASI’s test words | Good | Found the American Eagle Foundation logo once the .jpg file extension was removed, but none of the other specific files. The other search terms produced very good results |
2.3. Meta- Image Search Engines
Although a meta-search engine, which submits search requests to more than one other search engine, might seem to promise better results, this is seldom the case - especially when searching for images.
Meta-search engines offer fewer search options than their source search engines and the results tend to be slower and less reliable.
TASI evaluated nine meta-image search engines: five of them are general meta-search engines offering an image search option; two are dedicated to finding images; and two were really search ‘spring-boards’ rather than true meta-searches - they only allow the user to submit a search request to one other search engine at a time.
General meta- search engines
None of the general meta-engines tested performed very well.
Dogpile (http://www.dogpile.com/) draws all of its image results from FAST/ AllTheWeb, but it only presents the first nine images for any search - seldom enough to ensure a user’s needs are met. Dogpile does give the option of going to FAST for more results, which must be recommended, since FAST / AllTheWeb provides much richer metadata than Dogpile.
Both Excite (http://www.excite.com/) and Webcrawler (http://www.webcrawler.com/) draw on FAST / AllTheWeb and Ditto, presenting 9 from FAST and 12 from Ditto. Like Dogpile, users have to go to the source search engines to view more results. Excite and Webcrawler used to be independent, but are now owned by InfoSpace Inc (http://www.infospaceinc.com/), which accounts for the identical results.
Mamma (http://www.mamma.com/) draws on Picsearch and Lycos for its image search, presenting a combined total of 19 results and no option of getting any more. Since Lycos’s results always favour commercial Getty images (see above), these inevitably find their way into mamma’s results. Mamma adds its own commercial images - posters from Barewalls.com .
Fazzle (http://www.fazzle.com/) - which used to be called SearchOnline.info - is a little better than the others in this category. It supports Boolean and Phrase searching and draws its results from FAST, Lycos, Webshots and Ditto and throws in some additional advertising. Although it records the total number of hits, it only presents the ‘top’ hundred or so images. Many of these are dead links with missing images. It does enable the results to be sorted (by popularity, title, description, or domain), but it is not clear on what basis some of these rankings are made.
Specialised meta- search engines
Two meta- engines specialising in images were evaluated for this review, neither performed particularly well.
Ithaki (http://images.ithaki.net/) claims to search Ditto, Cobion, Altavista, FAST, Lycos and Excite, but TASI’s tests only prompted results from AllTheWeb/FAST and Ubbi (a Google-driven Spanish search engine). Ithaki only offers a simple search box, although states it supports Boolean and phrase searching. Despite the large indexes Ithaki searches, it delivers very low hit rates (less than 20) and includes many dead links.
March 2003 Update: After this review was published Ithaki contacted TASI with an update on their service. They put our limited results down to “an error in a perl expression in the Image search module” which had since been fixed - though they were still having some problems searching Altavista. We re-ran our tests and did acheive some very good results, averaging about 50 hits and drawn from a range of image search engines. Ithaki do not support Boolean for image searching since this is not always supported by the engines they search. They have had a help file available in the past, but this is currently “not available online since we’re re-writing it”.
Ixquick (http://www.ixquick.com/) searches FAST/AllTheWeb, GoGraph and Picsearch and delivers results from all of them. Although Ixquick reports large numbers of results, it is only possible to view a small selection of them: a search on ‘Elephant’ brought 14,000 hits, but Ixquick only presented 27 thumbnails. The quality of the results reflects the indexes Ixquick draws on - which translated into mixed results for TASI’s search terms. Despite its name, Ixquick was one of the slowest search engines encountered in this review.
‘Spring-board’ searching
These services are often classed as meta-search engines, but they are more like search ‘spring-boards’, since they only pass on the user’s search to one other search engine at a time (SearchEngineWatch term these ‘All-in-one’ search engines). Since they only offer very simple search options and are doing little more than passing a message on, they can seem like a waste of time and often are.
Photoseek (http://www.photoseek.net/) gives its users the option of searching AltaVista, ArtToday, Ditto, Excite, Lycos or WebSEEk. It simply passes on the request to the selected search engine and then frames the results. However, when tested it didn’t have its AltaVista search set up correctly (the results are Web sites rather than images) and it hasn’t registered that ArtToday has been renamed ClipArt or that WebSEEk is no longer functioning.
Search 22 Picture and Image Search Engines (http://www.search-22.com/images.html) offers a much better spring-board to image searching than Photoseek - or any of the proper meta- search engines listed above. Its twenty-two image search engines are a mixed bag, including Dogpile, Corbis, several clipart services and a collection of photos from Thailand. However, its first five engines are arguable the best places to search for general Web images: picsearch, Altavista, Ditto, Google and AllTheWeb. Although Search 22 does not offer the sophistication available when these are searched directly, it does enable the user to make a very quick check across a number of different services - useful in ascertaining which engines are worth searching in a more in-depth fashion.
3. Collection-based Image Search Engines
While image search engines strive for comprehensiveness and try as best they can to automatically identify and index images, collection-based engines draw on a smaller pool of catalogued images, usually from a database and usually indexed by humans. Large digital libraries and commercial photo or clip art providers are good examples of these collection-based search engines, though the largest group are probably adult-oriented services.
Digital library collections tend to be limited to particular subjects or formats (e.g. historic regional photographs or art collections), while the commercial stock photo collections aim for as broad a coverage as you would find in the independent image search engines evaluated above.
Because all of these collections are held in databases rather than embedded in Web pages, they tend to shut out the ‘spiders’ or ‘robots’ that independent search engines use to trawl and index the Internet. As a result, their images will seldom be found among general search engine results - unless, of course, the owner of the collection has entered into a commercial arrangement with a search engine service. There are some examples of this above (Lycos and Ditto).
While most commercial image collections are very happy to shut out search spiders, many owners of digital library collections would like to make their contents more easily discoverable. It seems likely that advances in metadata and Web-coding will improve this situation. Currently, any comprehensive search for images would need to include image search engines and the relevant collection-based engines. (For more on finding images, see TASI Resource Document: Searching the Internet for Existing Resources.)
Apart from their scope, the other chief difference between collection-based search engines and general image search engines, are that the collection-based engines usually search on metadata entered by humans, rather than automatically generated indexes that guess at their images’ contents. Consequently, the results from a collection-based search are likely to be much more consistent and relevant than those delivered by automated engines.
TASI’s database of Image Sites includes information on many image collections, particularly digital library collections. Since stock photo collections are aiming for same kind of scope and comprehensiveness as general image search engines, we evaluate four below to offer a comparison. In addition to two large commercial services, Corbis and Getty Images Creative, we consider WebShots, a large collaboratively built collection, and StopStock, which offers meta-search interface to several stock photo collections.
Corbis
http://www.corbis.com/
Founded by Microsoft’s Bill Gates in 1989, Corbis owns or licenses a vast collection of photography and fine art images. Corbis markets effectively to both professional and personals users: it has teamed with Yahoo to supply individuals with affordable images, and its own home page has two distinct search boxes, one marked ‘for professional use’ the other ‘for personal use.’ There is a great difference between these two searches - and not just the price of their resulting images. For this reason, we evaluate each search separately below.
Corbis Professional Search Engine
| Criteria | Overall Evaluation | Comments |
|---|---|---|
| Scope | Broad | Claims 2.1 million images |
| Search Options | Best | Simple and advanced searches. The advanced search is the best TASI saw in this review, enabling quite sophisticated filtering (before and after the search) and full Boolean, phrase and truncated searching |
| Performance | Very Good | Quick, numerous, relevant and high-quality results, as expected from a large stock photo database |
| Presentation | Good | Thumbnails first (up to 100 a page). Various file sizes are offered and the images are captioned and credited along with a date and keywords. There are clear statements of copyright and the larger images contain visible watermarks |
| Support | Good | Very full online help and the option of contacting Corbis by form or email |
| TASI’s test words | Good | Unsurprisingly, none of TASI’s specific files were found and there were poor results on the historic figure ‘Alexander Pope’. All the other terms tested produced very good results |
Corbis Personal Search Engine
| Criteria | Overall Evaluation | Comments |
|---|---|---|
| Scope | Unknown | Not stated, but clearly a smaller and inferior collection to the one the professional engine searches (above) |
| Search Options | Average | Simple search only, with little sophistication and no filtering before or after the search |
| Performance | Good | Results of a reasonable quality. Title and photographer are given and the option to buy 640x480 pixel images. Copyright is clearly stated and the larger images contain visible watermarks |
| Presentation | Good | Thumbnails first with links to larger images and online commerce facilities |
| Support | Basic | Basic online help and the ability to contact via a form |
| TASI’s test words | Average | Unsurprisingly, none of the specific files found but the results on the other terms were only slightly better than those found in some Web-indexing image search engines |
Getty Images Creative
http://creative.gettyimages.com/
Getty Images was founded in 1995 by Mark Getty and Jonathan Klein and has grown to rival Corbis in the commercial image market. Over the past few years it has bought or formed partnerships with many large image collections, including the Bridgeman Art Library, Hulton Archive, ImageBank, National Geographic and, very recently, the TimeLife collection. Many of Getty Images’ collections retain their own identities and often their own Web sites, but it is also possible to search across the collections via Getty Images’ Creative or News and Sport search engines. We evaluated Getty’s Creative search engine. Unlike Corbis’s search engine, Getty’s is clearly pitched at the professional user.
| Criteria | Overall Evaluation | Comments |
|---|---|---|
| Scope | Very broad | Very large and up-to-date collection, although statistics are unavailable |
| Search Options | Good | Simple and advanced options, including full Boolean, wildcard and phrase searching. Filtering is available before and after a search. Getty offers a potentially very useful feature for dealing with ambiguous search terms - a box which interrupts the search and requires users to choose from a number of possible meanings |
| Performance | Good | Generally very quick to search, though the number of hits was not always as high as expected, nor the quality as good |
| Presentation | Good | 12 thumbnails are returned as standard, but can be increased to 100. Information includes title, photographer and keywords, although the extent of the information varies from image to image depending on which collection it is drawn from. There is always a clear identification of copyright |
| Support | Good | Some help is available online and more is available via a form or phone number |
| TASI’s test words | Good | Unsurprisingly, none of the specific images were found. There were some good results on the general terms (elephant and flower), but poorer results on others - nothing for Alexander Pope, 4 for Cleopatra, and 144 for Madonna. Other Getty sites, such as News and Sport or the Hulton Archive, might be expected to produce better results on these. Getty’s clarification feature proved disappointing. The only alternatives offered for ‘Madonna’, for example, were ‘Sistine Madonna’ and ‘Virgin Mary’, while a search on ‘bank’ brought up ‘coin bank’ and ‘bank building’ but not river bank. The search on West Bank yielded no results |
Webshots
http://www.webshots.com/
Webshots is a community photo album with a commercial spin. Members store their own photographs online and these become available for others to search, view, and, in some cases, purchase.
| Criteria | Overall Evaluation | Comments |
|---|---|---|
| Scope | Eclectic | Claims 10 million photos - individual contributions supplemented with stock photo collections, making for uneven and eclectic subject coverage |
| Search Options | Average | Only offers a simple search interface. Exclude adult images |
| Performance | Average | Good results on some of the general terms, but poorer coverage of some of the more specific subjects searched on |
| Presentation | Average | Results are capped at 1000, with 12 thumbnails per page along with minimal additional information (pixel dimensions and ownership/copyright). Some Webshots images are available for purchase as screensaver images or as posters, others (the community contributions) can be viewed full-size or sent as e-cards. There are appropriate statements about copyright and use of images |
| Support | Poor | Limited online help and no easy way to ask questions about searching |
| TASI’s test words | Average | Surprisingly, the collection included the official American Eagle Foundation logo - someone had probably copied it and submitted it to Webshots. Mixed results on the other search terms - some very good results for general terms (‘elephant’ and ‘flower’), but poorer results for some of the specific terms (‘West Bank’ returned mostly archaeological artefacts) |
1StopStock
http://www.1stopstock.com/
1StopStock is a meta engine for stock photo collections. It searches 4 royalty-free collections (Comstock, Able Stock, Image State, and Digital Vision) and 4 rights-protected collections (Workbook Stock, Super Stock, Alamy and Comstock rights-protected).
| Criteria | Overall Evaluation | Comments |
|---|---|---|
| Scope | Unknown | Size and scope reflect the collections it searches. Seem broad but not vast |
| Search Options | Basic | Only offers a simple “keyword(s)” search and the choice of searching either royalty-free, rights-protected, or a single service |
| Performance | Mixed | Since 1StockShop is searching professional collections, the quality and relevancy are reasonably good, though vary from collection to collection. The Comstock searches were not functioning correctly when we checked: they simply opened up the Comstock homepage |
| Presentation | Mixed | If a single collection is being searched, 1StopStock will frame the results. If more than one is searched, each set of results pops up in a new window |
| Support | Average | Minimal online help available, but a form and email contact provided |
| TASI’s test words | Average | Good results on the general terms; mixed results on the more specific terms - only one gave adequate results for the ‘West Bank’ (the others returned pictures of buildings), one had Alexander Pope, and two had pictures to match ‘Dananus plexippus’ |
There are many stock photo collections available online. Here is a further selection:
| Brand X Pictures http://www.brandxpictures.com/ |
Magnum Photos http://www.magnumphotos.com/ |
| Bridgeman Art Library http://www.bridgeman.co.uk/ |
Mary Evans Picture Library http://www.mepl.co.uk/ |
| Clipart.com (was ArtToday) http://www.clipart.com/ |
National Geographic Images http://www.ngsimages.com/ |
| Christie’s Images http://www.christiesimages.com/ |
nonstock http://www.nonstock.com/ |
| Comstock http://www.comstock.com/ |
Photos To Go http://www.photostogo.com/ |
| eStock http://estockphoto.com/ |
Photospin http://www.photospin.com/ |
| Hulton Archive http://www.hultonarchive.com/ |
Stockbyte http://www.stockbyte.com/ |
| Index Stock Imagery http://www.indexstock.com/ |
TimePix http://www.timepix.com/ |
More can be found via the membership lists of the British Association of Picture Libraries and Agencies (BAPLA - http://www.bapla.org) and the Picture Archive Council of America (PACA - http://www.stockindustry.org/).
4. Content-based Search Engines
All the image search engines considered so far in this review have been based on text and context strategies or on associated catalogue entries. There have also been a number of attempts to build content-based search engines. Content-based image retrieval (CBIR) consider the characteristics of the image itself, for example its shapes and colours. To date, these attempts have been experimental and generally limited to individual collections.
Well publicised efforts have included Columbia University’s WebSEEk project (http://www.ctr.columbia.edu/webseek/ - no longer functioning) and IBM’s QBIC (Query by Image Content - http://wwwqbic.almaden.ibm.com/), which can be seen in action on the Hermitage Museum website (http://www.hermitagemuseum.org/).
The Institute for Image Data Research (IIDR) at Northumbria University (http://www.unn.ac.uk/iidr/) is currently working on a number of content-based retrieval projects and has a good set of links to other research programmes.
Some adult-site blockers have tried to use simple content analysis to filter out adult images, but these have not proved effective, since they are based on screening out images with particular (flesh-coloured) tonal values.
5. Conclusion
Image search engines attempt to give access to the wide range of images available on the Internet.
For those used to viewing well-indexed collections of quality images, the results of the large automated image search engines will probably disappoint. The poor quality of their offerings is not surprising, since they reflect the randomness and unevenness of the Web. The frequent irrelevancy of their results is also explicable, since the automated engines are guessing at their images’ visual subject content using indirect textual clues.
Anything, then, that enables the user to have more control over their image searching is helpful. The ability to filter a search - to include and exclude items - is important in any Web searching, but particularly so when searching for images. Many users will wish to exclude adult imagery from their search results, but it can also be very useful to limit by file type, file size, or colour - and the ability to use Boolean logic or phrases will greatly improve the relevancy of the results. An image search engine also needs to return a reasonable number of results, since in any given search a fair proportion of the images found are likely to be irrelevant or of insufficient quality.
Here the meta- image search engines fail their users. They deny them sophisticated filtered searching and generally only bring back a handful of the results they find. A meta- or spring-board service like Search22 can be useful in identifying the best image search engine for a particular task, but the final searching is better done directly, using the source search engines themselves. Google performed well in this review, but it is worth trying some of the others. As TASI’s evaluation suggestions, they each have strengths and weaknesses.
Collection-based image search engines include images selected for quality and indexed by hand. The images they contain are seldom found within the results of general Web search engines. Collection-based engines, then, will usually offer much better results than their search engine counterparts. The commercial and copyright issues will also be much clearer - although many users seem to prefer to look to image search engines for images they can ‘freely’ re-use, as if easy access and absent copyright notices lets them off the moral and legal hook. True copyright-free images are rare on the Web, and many image search engines do operate on commercial imperatives, even subtlety skewing their results towards commercial ends.
The Web is a fast-changing environment, so it is likely that some of the information in this review will quickly date. The reader is well advised to check the sites themselves and to ask the kind of questions listed in the evaluation criteria, above.
The most important question, however, is not on this list and can only be answered by each individual user: ‘For what purpose do I want to find images?’ The answer to this question needs to inform the choice of search tools, and the choice of search terms used.
For more advice on searching for images see TASI’s Advice Document: Finding Images Online.
Other sources of information include:
Search Engine Watch
http://www.searchenginewatch.com/
A site devoted to monitoring and evaluating search engine developments.
‘Image Search Engines’ by Richard Entlich (2001)
RLG DigiNews, vol. 5, no.6, December 15, 2001
http://www.rlg.org/preserv/diginews/diginews5-6.html
A review similar to this, conducted for the Research Libraries Group.
