main
Brett 2024-07-15 14:16:13 -04:00
parent 0c66fe85c6
commit 096c87cbf5
5 changed files with 152 additions and 74 deletions

3
.gitmodules vendored
View File

@ -10,3 +10,6 @@
[submodule "lib/stb"] [submodule "lib/stb"]
path = lib/stb path = lib/stb
url = https://github.com/nothings/stb url = https://github.com/nothings/stb
[submodule "lib/blt-graphics"]
path = lib/blt-graphics
url = https://git.tpgc.me/tri11paragon/BLT-With-Graphics-Template.git

View File

@ -1,5 +1,5 @@
cmake_minimum_required(VERSION 3.25) cmake_minimum_required(VERSION 3.25)
project(image-gp-6 VERSION 0.0.4) project(image-gp-6 VERSION 0.0.5)
include(FetchContent) include(FetchContent)
@ -10,6 +10,7 @@ option(ENABLE_TSAN "Enable the thread data race sanitizer" OFF)
set(CMAKE_CXX_STANDARD 17) set(CMAKE_CXX_STANDARD 17)
add_subdirectory(lib/blt-gp) add_subdirectory(lib/blt-gp)
add_subdirectory(lib/blt-graphics)
find_package( OpenCV REQUIRED ) find_package( OpenCV REQUIRED )
@ -25,7 +26,7 @@ add_executable(image-gp-6 ${PROJECT_BUILD_FILES})
target_compile_options(image-gp-6 PRIVATE -Wall -Wextra -Wpedantic -Wno-comment) target_compile_options(image-gp-6 PRIVATE -Wall -Wextra -Wpedantic -Wno-comment)
target_link_options(image-gp-6 PRIVATE -Wall -Wextra -Wpedantic -Wno-comment) target_link_options(image-gp-6 PRIVATE -Wall -Wextra -Wpedantic -Wno-comment)
target_link_libraries(image-gp-6 PRIVATE BLT blt-gp ${OpenCV_LIBS}) target_link_libraries(image-gp-6 PRIVATE BLT BLT_WITH_GRAPHICS blt-gp ${OpenCV_LIBS})
if (${ENABLE_ADDRSAN} MATCHES ON) if (${ENABLE_ADDRSAN} MATCHES ON)
target_compile_options(image-gp-6 PRIVATE -fsanitize=address) target_compile_options(image-gp-6 PRIVATE -fsanitize=address)

1
lib/blt-graphics Submodule

@ -0,0 +1 @@
Subproject commit 1470d8469d583f4a1b5aaf8916abbd4be2bba1ed

@ -1 +1 @@
Subproject commit 013ac3beddff3dbffafd5177e7972067cd2b5083 Subproject commit f7f20f39fe4f206c6f19e26ebfef7b261ee59ee4

View File

@ -15,11 +15,6 @@
* You should have received a copy of the GNU General Public License * You should have received a copy of the GNU General Public License
* along with this program. If not, see <https://www.gnu.org/licenses/>. * along with this program. If not, see <https://www.gnu.org/licenses/>.
*/ */
#define STB_IMAGE_RESIZE_IMPLEMENTATION
#define STB_IMAGE_IMPLEMENTATION
#define STB_IMAGE_WRITE_IMPLEMENTATION
#define STB_PERLIN_IMPLEMENTATION
#include <blt/gp/program.h> #include <blt/gp/program.h>
#include <blt/profiling/profiler_v2.h> #include <blt/profiling/profiler_v2.h>
#include <blt/gp/tree.h> #include <blt/gp/tree.h>
@ -29,8 +24,12 @@
#include <stb_image_resize2.h> #include <stb_image_resize2.h>
#include <stb_image_write.h> #include <stb_image_write.h>
#include <stb_perlin.h> #include <stb_perlin.h>
#include <blt/gfx/window.h>
#include "blt/gfx/renderer/resource_manager.h"
#include "blt/gfx/renderer/batch_2d_renderer.h"
#include "blt/gfx/renderer/camera.h"
#include <imgui.h>
#include "opencv2/imgcodecs.hpp" #include "opencv2/imgcodecs.hpp"
#include "opencv2/highgui.hpp"
#include "opencv2/imgproc.hpp" #include "opencv2/imgproc.hpp"
#include <random> #include <random>
@ -39,6 +38,11 @@ static constexpr long IMAGE_SIZE = 128;
static constexpr blt::size_t CHANNELS = 3; static constexpr blt::size_t CHANNELS = 3;
static constexpr blt::size_t DATA_SIZE = IMAGE_SIZE * IMAGE_SIZE; static constexpr blt::size_t DATA_SIZE = IMAGE_SIZE * IMAGE_SIZE;
blt::gfx::matrix_state_manager global_matrices;
blt::gfx::resource_manager resources;
blt::gfx::batch_renderer_2d renderer_2d(resources, global_matrices);
blt::gfx::first_person_camera_2d camera;
struct context struct context
{ {
float x, y; float x, y;
@ -46,34 +50,39 @@ struct context
struct image_t struct image_t
{ {
std::array<blt::u8, DATA_SIZE> image_data; std::array<blt::u8, DATA_SIZE> gray_data;
}; };
struct full_image_t struct full_image_t
{ {
std::array<blt::u8, DATA_SIZE * CHANNELS> image_data; std::array<blt::u8, DATA_SIZE * CHANNELS> rgb_data;
void load(const std::string& path) void load(const std::string& path)
{ {
int width, height, channels; int width, height, channels;
auto data = stbi_load(path.c_str(), &width, &height, &channels, CHANNELS); auto data = stbi_load(path.c_str(), &width, &height, &channels, CHANNELS);
stbir_resize_uint8_linear(data, width, height, 0, image_data.data(), IMAGE_SIZE, IMAGE_SIZE, 0, static_cast<stbir_pixel_layout>(CHANNELS)); stbir_resize_uint8_linear(data, width, height, 0, rgb_data.data(), IMAGE_SIZE, IMAGE_SIZE, 0, static_cast<stbir_pixel_layout>(CHANNELS));
stbi_image_free(data); stbi_image_free(data);
} }
void save(const std::string& str) void save(const std::string& str)
{ {
stbi_write_png(str.c_str(), IMAGE_SIZE, IMAGE_SIZE, CHANNELS, image_data.data(), 0); stbi_write_png(str.c_str(), IMAGE_SIZE, IMAGE_SIZE, CHANNELS, rgb_data.data(), 0);
} }
}; };
using fitness_data_t = std::array<image_t, 50>; //
//using fitness_data_t = std::array<image_t, 50>;
//
//fitness_data_t fitness_red;
//fitness_data_t fitness_green;
//fitness_data_t fitness_blue;
std::array<double, 64> fitness_values;
std::array<full_image_t, 64> generation_images;
std::vector<blt::gfx::texture_gl2D> gl_images;
fitness_data_t fitness_red;
fitness_data_t fitness_green;
fitness_data_t fitness_blue;
full_image_t base_data; full_image_t base_data;
full_image_t found_data; full_image_t found_data;
@ -100,7 +109,7 @@ blt::gp::prog_config_t config = blt::gp::prog_config_t()
.set_max_generations(50) .set_max_generations(50)
.set_mutation_chance(0.4) .set_mutation_chance(0.4)
.set_crossover_chance(0.9) .set_crossover_chance(0.9)
.set_pop_size(50) .set_pop_size(64)
.set_thread_count(0); .set_thread_count(0);
blt::gp::type_provider type_system; blt::gp::type_provider type_system;
@ -230,50 +239,53 @@ inline context get_ctx(blt::size_t i)
return ctx; return ctx;
} }
constexpr auto create_fitness_function(fitness_data_t& fitness_data, blt::size_t channel) constexpr auto create_fitness_function(blt::size_t channel)
{ {
return [&fitness_data, channel](blt::gp::tree_t& current_tree, blt::gp::fitness_t& fitness, blt::size_t in) { return [channel](blt::gp::tree_t& current_tree, blt::gp::fitness_t& fitness, blt::size_t in) {
auto& v = fitness_data[in]; auto& v = generation_images[in];
for (blt::size_t i = 0; i < DATA_SIZE; i++) for (blt::size_t i = 0; i < DATA_SIZE; i++)
{ {
context ctx = get_ctx(i); context ctx = get_ctx(i);
v.image_data[i] = static_cast<blt::u8>(current_tree.get_evaluation_value<float>(&ctx) * 255); v.rgb_data[i * CHANNELS + channel] = static_cast<blt::u8>(current_tree.get_evaluation_value<float>(&ctx) * 255);
auto dist = static_cast<float>(v.image_data[i]) - static_cast<float>(base_data.image_data[i * CHANNELS + channel]);
fitness.raw_fitness += std::sqrt(dist * dist);
} }
BLT_TRACE("Hello1"); fitness.raw_fitness = fitness_values[in];
cv::Mat img(IMAGE_SIZE, IMAGE_SIZE, CV_8UC3, v.image_data.data()); fitness.standardized_fitness = fitness.raw_fitness;
BLT_TRACE("Hello2"); fitness.adjusted_fitness = (1.0 / (1.0 + fitness.standardized_fitness));
cv::Mat img_hsv; // auto& v = fitness_data[in];
BLT_TRACE("Hello3"); // for (blt::size_t i = 0; i < DATA_SIZE; i++)
cv::cvtColor(img, img_hsv, cv::COLOR_RGB2HSV); // {
BLT_TRACE("Hello4"); // context ctx = get_ctx(i);
cv::Mat hist; // v.gray_data[i] = static_cast<blt::u8>(current_tree.get_evaluation_value<float>(&ctx) * 255);
BLT_TRACE("Hello5"); //
cv::calcHist(&img_hsv, 1, channels, cv::Mat(), hist, 2, histSize, ranges, true, false); // auto dist = static_cast<float>(v.gray_data[i]) - static_cast<float>(base_data.rgb_data[i * CHANNELS + channel]);
BLT_TRACE("Hello6"); //
cv::normalize(hist, hist, 0, 1, cv::NORM_MINMAX, -1, cv::Mat()); // fitness.raw_fitness += std::sqrt(dist * dist);
BLT_TRACE("Hello7"); // }
// cv::Mat img(IMAGE_SIZE, IMAGE_SIZE, CV_8UC1, v.gray_data.data());
// cv::Mat img_rgb;
// cv::Mat img_hsv;
// cv::cvtColor(img, img_rgb, cv::COLOR_GRAY2RGB);
// cv::cvtColor(img_rgb, img_hsv, cv::COLOR_RGB2HSV);
// cv::Mat hist;
// cv::calcHist(&img_hsv, 1, channels, cv::Mat(), hist, 2, histSize, ranges, true, false);
// cv::normalize(hist, hist, 0, 1, cv::NORM_MINMAX, -1, cv::Mat());
//
// auto comp = 1.0 - cv::compareHist(base_image_hist, hist, cv::HISTCMP_CHISQR);
//
// fitness.raw_fitness *= comp;
auto comp = cv::compareHist(base_image_hist, hist, cv::HISTCMP_CORREL); // fitness.standardized_fitness = fitness.raw_fitness / IMAGE_SIZE;
// fitness.adjusted_fitness = (1.0 / (1.0 + fitness.standardized_fitness));
fitness.standardized_fitness = fitness.raw_fitness / IMAGE_SIZE;
fitness.adjusted_fitness = (1.0 / (1.0 + fitness.standardized_fitness)) * comp;
}; };
} }
constexpr auto fitness_function_red = create_fitness_function(fitness_red, 0); constexpr auto fitness_function_red = create_fitness_function(0);
constexpr auto fitness_function_green = create_fitness_function(fitness_green, 1); constexpr auto fitness_function_green = create_fitness_function(1);
constexpr auto fitness_function_blue = create_fitness_function(fitness_blue, 2); constexpr auto fitness_function_blue = create_fitness_function(2);
void evaluate_program(blt::gp::gp_program& program) void execute_generation(blt::gp::gp_program& program)
{
BLT_DEBUG("Begin Generation Loop");
while (!program.should_terminate())
{ {
BLT_TRACE("------------{Begin Generation %ld}------------", program.get_current_generation()); BLT_TRACE("------------{Begin Generation %ld}------------", program.get_current_generation());
BLT_START_INTERVAL("Image Test", "Gen"); BLT_START_INTERVAL("Image Test", "Gen");
@ -288,6 +300,14 @@ void evaluate_program(blt::gp::gp_program& program)
BLT_TRACE("----------------------------------------------"); BLT_TRACE("----------------------------------------------");
std::cout << std::endl; std::cout << std::endl;
} }
void evaluate_program(blt::gp::gp_program& program)
{
BLT_DEBUG("Begin Generation Loop");
while (!program.should_terminate())
{
execute_generation(program);
}
} }
void print_stats(blt::gp::gp_program& program) void print_stats(blt::gp::gp_program& program)
@ -332,9 +352,9 @@ void write_tree(blt::size_t index, blt::size_t best_red, blt::size_t best_blue,
for (blt::size_t i = 0; i < DATA_SIZE; i++) for (blt::size_t i = 0; i < DATA_SIZE; i++)
{ {
found_data.image_data[i * CHANNELS] = fitness_red[best_red].image_data[i]; found_data.rgb_data[i * CHANNELS] = generation_images[best_red].rgb_data[i * CHANNELS];
found_data.image_data[i * CHANNELS + 1] = fitness_green[best_green].image_data[i]; found_data.rgb_data[i * CHANNELS + 1] = generation_images[best_green].rgb_data[i * CHANNELS + 1];
found_data.image_data[i * CHANNELS + 2] = fitness_blue[best_blue].image_data[i]; found_data.rgb_data[i * CHANNELS + 2] = generation_images[best_blue].rgb_data[i * CHANNELS + 2];
} }
found_data.save("best_image_" + std::to_string(index) + ".png"); found_data.save("best_image_" + std::to_string(index) + ".png");
@ -386,15 +406,23 @@ void make_operator_image(T op, Args... args)
stbi_write_png((blt::type_string<T> + ".png").c_str(), IMAGE_SIZE, IMAGE_SIZE, CHANNELS, value.get(), 0); stbi_write_png((blt::type_string<T> + ".png").c_str(), IMAGE_SIZE, IMAGE_SIZE, CHANNELS, value.get(), 0);
} }
int main() void init(const blt::gfx::window_data&)
{ {
using namespace blt::gfx;
for (blt::size_t i = 0; i < config.population_size; i++)
{
gl_images.emplace_back(IMAGE_SIZE, IMAGE_SIZE, GL_RGB8);
resources.set(std::to_string(i), &gl_images.back());
}
BLT_INFO("Starting BLT-GP Image Test"); BLT_INFO("Starting BLT-GP Image Test");
BLT_INFO("Using Seed: %ld", SEED); BLT_INFO("Using Seed: %ld", SEED);
BLT_START_INTERVAL("Image Test", "Main"); BLT_START_INTERVAL("Image Test", "Main");
BLT_DEBUG("Setup Base Image"); BLT_DEBUG("Setup Base Image");
base_data.load("../Rolex_De_Grande_-_Joo.png"); base_data.load("../Rolex_De_Grande_-_Joo.png");
cv::Mat base_image_mat{IMAGE_SIZE, IMAGE_SIZE, CV_8UC3, base_data.image_data.data()}; cv::Mat base_image_mat{IMAGE_SIZE, IMAGE_SIZE, CV_8UC3, base_data.rgb_data.data()};
cv::cvtColor(base_image_mat, base_image_hsv, cv::COLOR_RGB2HSV); cv::cvtColor(base_image_mat, base_image_hsv, cv::COLOR_RGB2HSV);
cv::calcHist(&base_image_hsv, 1, channels, cv::Mat(), base_image_hist, 2, histSize, ranges, true, false); cv::calcHist(&base_image_hsv, 1, channels, cv::Mat(), base_image_hist, 2, histSize, ranges, true, false);
@ -412,9 +440,54 @@ int main()
program_green.generate_population(type_system.get_type<float>().id(), fitness_function_green); program_green.generate_population(type_system.get_type<float>().id(), fitness_function_green);
program_blue.generate_population(type_system.get_type<float>().id(), fitness_function_blue); program_blue.generate_population(type_system.get_type<float>().id(), fitness_function_blue);
evaluate_program(program_red); // evaluate_program(program_red);
evaluate_program(program_green); // evaluate_program(program_green);
evaluate_program(program_blue); // evaluate_program(program_blue);
global_matrices.create_internals();
resources.load_resources();
renderer_2d.create();
}
void update(const blt::gfx::window_data& data)
{
global_matrices.update_perspectives(data.width, data.height, 90, 0.1, 2000);
camera.update();
camera.update_view(global_matrices);
global_matrices.update();
for (blt::size_t i = 0; i < config.population_size; i++)
gl_images[i].upload(generation_images[i].rgb_data.data(), IMAGE_SIZE, IMAGE_SIZE, GL_RGB);
if (ImGui::Begin("Program Control"))
{
ImGui::Button("Run Generation");
if (ImGui::IsItemClicked())
{
execute_generation(program_red);
execute_generation(program_green);
execute_generation(program_blue);
}
ImGui::End();
}
for (blt::size_t i = 0; i < config.population_size; i++)
{
renderer_2d.drawRectangleInternal(std::to_string(i),
{static_cast<float>(IMAGE_SIZE * i), static_cast<float>(IMAGE_SIZE * i), IMAGE_SIZE, IMAGE_SIZE});
}
renderer_2d.render(data.width, data.height);
}
int main()
{
blt::gfx::init(blt::gfx::window_data{"My Sexy Window", init, update}.setSyncInterval(1));
global_matrices.cleanup();
resources.cleanup();
renderer_2d.cleanup();
blt::gfx::cleanup();
BLT_END_INTERVAL("Image Test", "Main"); BLT_END_INTERVAL("Image Test", "Main");