1

我正在尝试从缓冲区中读取 10 位图像并使用 boost::gil 对其进行分析。我注意到存在 rgb8_image_t 类型， gray8_image_t 类型，但找不到 gray10_image_t 类型（这是我需要的）。这在gil中存在吗？谢谢！

编辑：使用这些类型定义时：

// reference type
typedef boost::gil::bit_aligned_pixel_reference<unsigned short, boost::mpl::vector1_c<unsigned, 10>, boost::gil::gray_layout_t, true> gray10_ref_t;

// iterator type
typedef boost::gil::bit_aligned_pixel_iterator<gray10_ref_t> gray10_ptr_t;

// pixel type
typedef std::iterator_traits<gray10_ptr_t>::value_type gray10_pixel_t;

// pixel storage to read, contains 3 10-bit gray pixels, all with value of 1 as per the following layout
// spaces show byte breaks, bars show pixel breaks, lsb first
// 10000000 00|100000 0000|1000 00000000
std::uint8_t data[4] = { 0x01, 0x04, 0x10, 0x00 };

// an iterator to the start of the storage
gray10_ptr_t p(&data[0], 0);

// check the expected pixel values
assert(p[0] == 0x01);
assert(p[1] == 0x01);
assert(p[2] == 0x01);

编译时出现以下错误：

untitled.cpp:17:6: error: ‘uint8_t’ in namespace ‘std’ does not name a type 
std::uint8_t data[4] = { 0x01, 0x04, 0x10, 0x00 };
     ^
untitled.cpp:20:17: error: ‘data’ was not declared in this scope
 gray10_ptr_t p(&data[0], 0);
                 ^
In file included from /usr/include/boost/predef/detail/_cassert.h:14:0,
                 from /usr/include/boost/predef/library/c/_prefix.h:11,
                 from /usr/include/boost/predef/library/c.h:11,
                 from /usr/include/boost/predef/library.h:11,
                 from /usr/include/boost/predef.h:14,
                 from /usr/include/boost/smart_ptr/detail/yield_k.hpp:28,
                 from /usr/include/boost/smart_ptr/detail/spinlock_sync.hpp:18,
                 from /usr/include/boost/smart_ptr/detail/spinlock.hpp:50,
                 from /usr/include/boost/smart_ptr/detail/spinlock_pool.hpp:25,
                 from /usr/include/boost/smart_ptr/shared_ptr.hpp:34,
                 from /usr/include/boost/shared_ptr.hpp:17,
                 from /usr/include/boost/gil/extension/io/io_error.hpp:23,
                 from /usr/include/boost/gil/extension/io/tiff_io.hpp:29,
                 from untitled.cpp:1:
untitled.cpp:23:13: error: expected ‘)’ before ‘==’ token
 assert(p[0] == 0x01);
             ^
untitled.cpp:23:13: error: expected ‘)’ before ‘==’ token
untitled.cpp:24:13: error: expected ‘)’ before ‘==’ token
 assert(p[1] == 0x01);
             ^
untitled.cpp:24:13: error: expected ‘)’ before ‘==’ token
untitled.cpp:25:13: error: expected ‘)’ before ‘==’ token
 assert(p[2] == 0x01);
             ^
untitled.cpp:25:13: error: expected ‘)’ before ‘==’ token

有谁知道如何解决这些问题？谢谢。

4

1 回答 1

2

没有，但您可以自己定义一个。

由于您的像素类型不是字节对齐的，因此您需要声明一个位对齐的像素和相关类型：

位对齐像素（和图像）比打包像素更复杂。由于打包像素是字节对齐的，我们可以使用 C++ 引用作为打包像素的引用类型，并将 C 指针用作一行打包像素的 x_iterator。对于位对齐结构，我们需要一个特殊的参考代理类（bit_aligned_pixel_reference）和迭代器类（bit_aligned_pixel_iterator）。位对齐像素的值类型是packed_pixel。

一个工作示例：

#include <boost/gil.hpp>
#include <boost/mpl/vector.hpp>
#include <cstdint>

int main()
{
    // reference type
    using gray10_ref_t = boost::gil::bit_aligned_pixel_reference<unsigned short, boost::mpl::vector1_c<unsigned, 10>, boost::gil::gray_layout_t, true>;
    // iterator type
    using gray10_ptr_t = boost::gil::bit_aligned_pixel_iterator<gray10_ref_t>;
    // pixel type
    using gray10_pixel_t = std::iterator_traits<gray10_ptr_t>::value_type;

    // pixel storage to read, contains 3 10-bit gray pixels, all with value of 1 as per the following layout
    // spaces show byte breaks, bars show pixel breaks, lsb first
    // 10000000 00|100000 0000|1000 00000000
    std::uint8_t data[4] = { 0x01, 0x04, 0x10, 0x00 };

    // an iterator to the start of the storage
    gray10_ptr_t p(&data[0], 0);

    // check the expected pixel values
    assert(p[0] == 0x01);
    assert(p[1] == 0x01);
    assert(p[2] == 0x01);
}
于 2018-07-29T23:44:58.733 回答