OpenCV中GPU模块 CUDA 函数

分享一下我老师大神的人工智能教程！零基础，通俗易懂！http://blog.csdn.net/jiangjunshow

也欢迎大家转载本篇文章。分享知识，造福人民，实现我们中华民族伟大复兴！

               

The OpenCV GPU module is a set of classes and functions to utilize GPU computational capabilities. It is implemented using NVIDIA* CUDA* Runtime API and supports only NVIDIA GPUs.

 

1.      getCudaEnableDeviceCount：returns the number of installed CUDA-enabled devices；

2.      setDevice：sets adevice and initializes it for the current thread；

3.      getDevice：returns the current device index set or initialized by default；

4.      resetDevice：explicitly destroys and cleans up all resources associated with the current device in the current process；

5.      FeatureSet：enumeration providing GPU computing features；

6.      class::TargetArchs：class providing a set of static methods to check what NVIDIA card architecture the GPU module was built for；

7.      class::DeviceInfo：class providing functionality for querying the specified GPU properties；

8.      DeviceInfo::name：returns the device name；

9.      DeviceInfo::majorVersion：returns the major compute capability version；

10.  DeviceInfo::minorVersion：returns the minor compute capability version；

11.  DeviceInfo::multiProcessorCount：returns the number of streaming multiprocessors；

12.  DeviceInfo::freeMemory：returns the amount of free memory in bytes；

13.  DeviceInfo::totalMemory：returns the amount of total memory in bytes；

14.  DeviceInfo::supports：provides information on GPU feature support；

15.  DeviceInfo::isCompatible：checks the GPU module and device compatibility；

16.  DeviceInfo::deviceID：returns system index of the GPU device starting with 0；

17.  struct::PtrStepSz：lightweight class encapsulating pitched memory on a GPU and passed to nvcc-compiled code(CUDA kernels)；

18.  struct::PtrStep：structure similar to gpu::PtrStepSz but containing only a pointer and row step；

19.  class::GpuMat：base storage class for GPU memory with reference counting. Its interface matches the Mat interface；

20.  createContinuous：creates a continuous matrix in the GPU memory；

21.  ensureSizeIsEnough：ensures that the size of a matrix is big enough and the matrix has a proper type；

22.  registerPageLocked：page-locks the memory of matrix and maps it for the device(s)；

23.  unregisterPageLocked：unmaps the memory of matrix and makes it pageable again；

24.  class::CudaMem：class with reference counting wrapping special memory type allocation functions from CUDA. Its interface is also Mat()-like but with additional memory type parameters；

25.  CudaMem::createMatHeader：creates a header without reference counting to gpu::CudaMem data；

26.  CudaMem::createGpuMatHeader：maps CPU memory to GPU address space and create the gpu::GpuMat header without reference counting for it；

27.  CudaMem::canMapHostMemory：returns true if the current  hardware supports address space mapping and ALLOC_ZEROCOPY memory allocation；

28.  class::Stream：this class encapsulates a queue of asynchronous calls. Some functions have overloads with the additional gpu::Stream parameter；

29.  Stream::queryIfComplete：returns true if the current stream queue is finished, otherwise, it returns false；

30.  Stream::waitForCompletion：blocks the current CPU thread until all operations in the stream are complete；

31.  Stream::enqueueDownload：copies data from device to host；

32.  Stream::enqueueUpload：copies data from host to device；

33.  Stream::enqueueCopy：copies data from device to device；

34.  Stream::enqueueMemSet：initializes or sets device memory to a value；

35.  Stream::enqueueConvert：converts matrix type, ex from float to uchar depending on type；

36.  Stream::enqueueHostCallback：adds a callback to be called on the host after all currently enqueued items in the stream have completed；

37.  struct::StreamAccessor：class that enables getting cuda Stream_t from gpu::Stream and is declared in stream_accessor.hpp because it is the only public header that depends on the CUDA Runtime API；

38.  gemm(cv::gemm)：performs generalized matrix multiplication；

39.  transpose(cv::transpose)：transpose a matrix；

40.  flip(cv::flip)：flips a 2D matrix around vertical, horizontal, or both axes；

41.  LUT(cv::LUT)：transforms the source matrix into the destination matrix using the given look-up table:dst(I) = lut(src(I))；

42.  merge(cv::merge)：makes a multi-channel matrix out of several single-channel matrices；

43.  split(cv::split)：copies each plane of a multi-channel matrix into an array；

44.  magnitude(cv::magnitude)：computes magnitudes of complex matrix elements；

45.  magnitudeSqr：computes squared magnitudes of complex matrix elements；

46.  phase(cv::phase)：computes polar angles of complex matrix elements；

47.  cartToPolar(cv::cartToPolar)：converts Cartesian coordinates into polar；

48.  polarToCart(cv::polarToCart)：converts polar coordinates into Cartesian；

49.  normalize(cv::normalize)：normalizes the norm or value range of an array；

50.  add(cv::add)：computes a matrix-matrix or matrix-scalar sum；

51.  subtract(cv::subtract)：computes a matrix-matrix or matrix-scalar difference；

52.  multiply(cv::multiply)：computes a matrix-matrix or matrix-scalar per-element product；

53.  divide(cv::divide)：computes a matrix-matrix or matrix-scalar division；

54.  addWeighted(cv::addWeighted)：computes the weighted sum of two arrays；

55.  abs(cv::abs)：computes an absolute value of each matrix element；

56.  sqr：computes a square value of each matrix element；

57.  sqrt(cv::sqrt)：computes a square root of each matrix element；

58.  exp(cv::exp)：computes an exponent of each matrix element；

59.  log(cv::log)：computes a natural logarithm of absolute value of each matrix element；

60.  pow(cv::pow)：raises every matrix elem