我想知道如何做，就像我在标题中所说的那样：
我想要对一些对象进行计数，从便携式设备（例如 iPhone 或 Android 手机）的相机中读取图像，
我只需要两个特定功能。
- 识别并计算物体的数量
- 识别物体的颜色（这样我就可以计算每种颜色有多少个）。
一个非常简单的例子。
我有一堆乐高积木，它们的尺寸都相同。我知道它们总是水平对齐，有时它们不是垂直对齐。我需要数一下我有多少种颜色。
我知道我有相同尺寸的作品，只是颜色发生了变化。
我认为只有 10 种颜色可用。
我可以详细说明图像（例如模糊和其他内容），但我不知道如何读取我有多少块。
你能告诉我一些如何做的想法（以及在 iOS 和 Android 上使用什么样的库 - 首先是 Android -）或者一些出版物（免费的 pdf 或书籍，甚至出版的书籍，即使它们不是免费的）教如何做从图像中读取数据。
该程序应该起到相同的作用：
我启动该程序，当程序识别出它正在查看（使用集成凸轮）一些特定对象时，它会拍摄一张照片并对其进行详细说明，告诉我每种颜色有多少。
提前致谢，任何形式的帮助都会
I am wondering how to do as I said in the title:
I want to have some objects counted reading an image from the camera of the portable device (such as iPhone or Android phones)
I need only two specific functions.
- Recognize and count the amount of objects
- Recognize the color of the object (so I can count how many of each color I have).
A very simple example.
I have a stack of pieces of LEGO, all of them the same dimensions. I know they always will be aligned horizontaly, sometimes they are not verically aligned. I need to count how many of each colour I have.
I know that I have pieces with the same dimensions, only the colour change.
I have i think only 10 colour avaible.
I can elaborate the image (such as blur and other stuff) but I don't know how to read how many pieces I have.
Can you tell me some Ideas how to do (and what kind of libraries to use both for iOS and Android -android first-) or maybe some publication (free pdf or books or even publicated books even if they're not free) teaching how to read data from images.
The program should be act as the same:
I start the program, when the program recognize it is looking at (using the integrated cam) some specific objects, ittake a picture and elaborate it, telling how much of each color I have.
Thanks in advance, ANY kind of help will be
发布评论
评论（1）
我承认距离我上次涉足计算机视觉已经有 10 年了，但当时我使用了 OpenCV 库，这些库似乎仍然很强大，并且支持 Android：
http://opencv.willowgarage.com/wiki/Android
和 iOS：
http://www.eosgarden.com/en/opensource/opencv-ios/overview/
I'll admit it is 10 years since I last dabbled with computer vision, but back then I used the OpenCV libraries, and these still seem to be going strong, and support on Android:
http://opencv.willowgarage.com/wiki/Android
and iOS:
http://www.eosgarden.com/en/opensource/opencv-ios/overview/