It is possible but there are some limitations. Using the canvas element it is feasible to access the pixel information of images. However gQ isn't implemented with the canvas but with plain DOM.
It means that you have to find another way to pass the pixel information to javascript. The most obvious one is to use a bi-dimensional array. You will fill the array with a value to indicate a value for each pixel of the image. This value could be true/false to differentiate between transparent and plain pixels. It could even have more value to implement more subtle mechanism (like you would need for a fighting game) .
This has some drawback. First you have to create all those arrays (you could do it automatically with a server side or offline script). Then those arrays will take a lot of place in memory. Finally if your sprite are animated you may want to take this into consideration.
In some situation however you really need per-pixel collision and this is a solution that will work on all the browser where gQ works.
If you'r prepared to drop support for older version of IE you could use the canvas element to read the pixel value. This is create cleaner code, remove the need for a separate array and take less memory.
SInce both of those solution will require quite a lot of work I would consider using an engine that already implement this.
Comments
It means that you have to find another way to pass the pixel information to javascript. The most obvious one is to use a bi-dimensional array. You will fill the array with a value to indicate a value for each pixel of the image. This value could be true/false to differentiate between transparent and plain pixels. It could even have more value to implement more subtle mechanism (like you would need for a fighting game) .
This has some drawback. First you have to create all those arrays (you could do it automatically with a server side or offline script). Then those arrays will take a lot of place in memory. Finally if your sprite are animated you may want to take this into consideration.
In some situation however you really need per-pixel collision and this is a solution that will work on all the browser where gQ works.
If you'r prepared to drop support for older version of IE you could use the canvas element to read the pixel value. This is create cleaner code, remove the need for a separate array and take less memory.
SInce both of those solution will require quite a lot of work I would consider using an engine that already implement this.