android camera

http://stackoverflow.com/questions/15578614/camera-setfocusareas-not-working-android-4-0

public final void autoFocus (Camera.AutoFocusCallback cb)
So you should be able to add a few lines to your code:

Rect newRect = new Rect(left,top,right, bottom);
Camera.Parameters params = mCamera.mCameraInstance.getParameters();
Camera.Area focusArea = new Camera.Area(newRect, 1000);
List<Camera.Area> focusAreas = new ArrayList<Camera.Area>();
focusAreas.add(focusArea);
params.setFocusAreas(focusAreas);

/**
 * ISO changes occurs passively after setting it in the cameraParams. Auto focus
 * parameters that have been set don't take affect until calling Camera.autoFocus()
 */
mCamera.autoFocus(new Camera.AutoFocusCallback()
{
    @Override
    public void onAutoFocus(boolean success, Camera camera)
    {
        Log.d(TAG, "onAutoFocus() " + success);
    }
});

http://androiddev.orkitra.com/?p=71685


ollowing several tutorials and examples I came up with the next algorithm to set the camera focus on a specific spot, the problem is that the camera completely ignores the spot and performs a normal overall focus instead of the rect area which I have specified. Is there anything else that I am missing in the algorithm? This has been tested on several phones all with Android 4.0 and above, so the focus area API is supported on these devices. Note, the app I am writing works in landscape mode only.

@Override
public boolean onTouchEvent(final MotionEvent event)
{
    if (event.getAction() == MotionEvent.ACTION_UP)
    {
        final int screenX = (int)event.getRawX();
        final int screenY = (int)event.getRawY();

        this.submitFocusAreaRect(screenX, screenY);
    }
}

private void submitFocusAreaRect(final int X, final int Y)
{
    Camera.Parameters cameraParameters = camera.getParameters();

    if (cameraParameters.getMaxNumFocusAreas() == 0)
    {
        return;
    }

    // Calculate the focus spot area

    Rect areaFocusRect = new Rect();
    areaFocusRect.set(0, 0, cameraSurfaceView.getWidth(), cameraSurfaceView.getHeight());
    areaFocusRect.top += (cameraSurfaceView.getHeight() - cameraSurfaceView.getWidth()) / 2;
    areaFocusRect.bottom -= areaFocusRect.top;

    int i = X - areaFocusRect.left;
    int j = Y - areaFocusRect.top;
    int k = (int)Math.min(900.0F, Math.max(-900.0F, -1000.0F + 2000.0F * (i / areaFocusRect.width())));
    int m = (int)Math.min(900.0F, Math.max(-900.0F, -1000.0F + 2000.0F * (j / areaFocusRect.height())));
    areaFocusRect.set(k - 100, m - 100, k + 100, m + 100);

    // Submit focus area to camera

    ArrayList<Camera.Area> focusAreas = new ArrayList<Camera.Area>();
    focusAreas.add(new Camera.Area(areaFocusRect, 1));

    cameraParameters.setFocusAreas(focusAreas);
    cameraParameters.setFocusMode(Camera.Parameters.FOCUS_MODE_AUTO);
    camera.setParameters(cameraParameters);

    // Start the autofocus operation

    camera.autoFocus(this);
}

 

Answer
before cameraParameters.setFocusAreas(focusAreas);, you should add this:

cameraParameters.setFocusMode(Camera.Parameters.FOCUS_MODE_AUTO);


深度学习是机器学习的一个子领域,它基于人工神经网络的研究,特别是利用多层次的神经网络来进行学习和模式识别。深度学习模型能够学习数据的高层次特征,这些特征对于图像和语音识别、自然语言处理、医学图像分析等应用至关重要。以下是深度学习的一些关键概念和组成部分: 1. **神经网络(Neural Networks)**:深度学习的基础是人工神经网络,它是由多个层组成的网络结构,包括输入层、隐藏层和输出层。每个层由多个神经元组成,神经元之间通过权重连接。 2. **前馈神经网络(Feedforward Neural Networks)**:这是最常见的神经网络类型,信息从输入层流向隐藏层,最终到达输出层。 3. **卷积神经网络(Convolutional Neural Networks, CNNs)**:这种网络特别适合处理具有网格结构的数据,如图像。它们使用卷积层来提取图像的特征。 4. **循环神经网络(Recurrent Neural Networks, RNNs)**:这种网络能够处理序列数据,如时间序列或自然语言,因为它们具有记忆功能,能够捕捉数据中的时间依赖性。 5. **长短期记忆网络(Long Short-Term Memory, LSTM)**:LSTM 是一种特殊的 RNN,它能够学习长期依赖关系,非常适合复杂的序列预测任务。 6. **生成对抗网络(Generative Adversarial Networks, GANs)**:由两个网络组成,一个生成器和一个判别器,它们相互竞争,生成器生成数据,判别器评估数据的真实性。 7. **深度学习框架**:如 TensorFlow、Keras、PyTorch 等,这些框架提供了构建、训练和部署深度学习模型的工具和库。 8. **激活函数(Activation Functions)**:如 ReLU、Sigmoid、Tanh 等,它们在神经网络中用于添加非线性,使得网络能够学习复杂的函数。 9. **损失函数(Loss Functions)**:用于评估模型的预测与真实值之间的差异,常见的损失函数包括均方误差(MSE)、交叉熵(Cross-Entropy)等。 10. **优化算法(Optimization Algorithms)**:如梯度下降(Gradient Descent)、随机梯度下降(SGD)、Adam 等,用于更新网络权重,以最小化损失函数。 11. **正则化(Regularization)**:技术如 Dropout、L1/L2 正则化等,用于防止模型过拟合。 12. **迁移学习(Transfer Learning)**:利用在一个任务上训练好的模型来提高另一个相关任务的性能。 深度学习在许多领域都取得了显著的成就,但它也面临着一些挑战,如对大量数据的依赖、模型的解释性差、计算资源消耗大等。研究人员正在不断探索新的方法来解决这些问题。
评论
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值