An artificial intelligent application in order to detect breeds of cats


Ml5.js is a library which provide access to machine learning algorithms and models in the browser.

We give to a computer many images, like cat's images, but with different cat in each image.




Artificial intelligence is able to recognize images especially images of cats.


these videos explains how it works











mobilenet tells you what the AI recognizes and also the pourcent of confidence.


As I explain to you before with mlc.5 and with MobileNet, the AI can recognize an image of cat based on a image database.Indeed, in this database we can find many diverses images of cats of many different breeds.




You need to teach to the AI many different images of what you want it to recognize.
Therefore, for an application which can detect breeds of cats, you need to teach to the AI many images of breeds of cats.




As of 2019, The International Cat Association (TICA) recognizes 71 standardized breeds,the Cat Fanciers' Association (CFA) recognizes 44, and Fédération Internationale Féline (FIFe) recognizes 43. Inconsistency in breed classification and naming among registries means that an individual animal may be considered different breeds by different registries (though not necessarily eligible for registry in them all, depending on its exact ancestry). For example, TICA's Himalayan is considered a colorpoint variety of the Persian by CFA, while the Javanese (or Colorpoint Longhair) is a color variation of the Balinese in TICA and CFA; both breeds are merged (along with the Colorpoint Shorthair) into a single "mega-breed", the Colourpoint, by the World Cat Federation (WCF), who have repurposed the name "Javanese" for the Oriental Longhair. Also, "Colo[u]rpoint Longhair" refers to different breeds in other registries. There are many examples of nomenclatural overlap and differences of this sort. Furthermore, many geographical and cultural names for cat breeds are fanciful selections made by Western breeders to be exotic sounding and bear no relationship to the actual origin of the breeds; the Balinese, Javanese, and Himalayan are all examples of this trend.


To be sure of the breed of one cat, you should check with a veteritarian. The more images the Artificial Intelligence encounters, the more raliable it is.




Once you have find an image of cat, you can bring it in the application in order to be recognized. And the application will work.