andy,
i would like to make a software, who can learn if a .jpg
of a human is male or female. the software should use a
neuronal network. the idea is, that first i view some
hundret pictures of people, and always click male or
female, then, the network tryes to learn, and should be
able to decide if a picture is male or female. does
anybody can give me tips, how i can do this?
is there code available for the neuronal network things?
i would be very happy for input!
I was the lead architect and developer for the KnowledgeScape software
(
www.kscape.com) that includes smart objects, expert systems, neural
networks, and genetic algorithms for real-time process control. It also
contains some image analysis capabilities.
The mathematics behind a neural network are quite simple to code. This will
probably not be much of a challenge for you. If you need help, there are
thousands of books on the subject. You are bound to have free access to
some of these at your local library, local university or book stores. You
can get a bunch of resources from the internet also.
The challenge is going to be deciding how to encode the features in the
image that can be used to distinguish the sex of the person in the image.
Basically, once the image is created, you will be left with a
two-dimensional array of RGB color values. I am not suggesting that your
task is impossible. But I am suggesting that the biggest challenge you will
"face" (pardon the pun) is to record the images consistently and present
them in a way that provides the functional relationship between the sex of
the subject and the encoded features.
This will also lead you into a decision about the type of neural network
that you use.
So, I suggest that you find a good book on programming a neural-network,
create some code, and then work on the encoding of the image to include the
features that identify whether the subject is male or female.
Hope that helps.
Regards,
Randy
