Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Improve performance of Material Detector #459

Merged
merged 4 commits into from
Jul 17, 2024
Merged

Improve performance of Material Detector #459

Show file tree
Hide file tree
Changes from 3 commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
fd3caf7
adjust material detector to use waterfill OCR instead of straight OCR.
jw098 Jul 9, 2024
3363a43
clean up old code
jw098 Jul 9, 2024
60e1c4d
parallelize detection of material value
jw098 Jul 9, 2024
7ac408e
adjust threshold for detecting dark vs light text
jw098 Jul 13, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
42 changes: 41 additions & 1 deletion SerialPrograms/Source/NintendoSwitch/DevPrograms/TestProgramSwitch.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -267,12 +267,52 @@ void TestProgram::program(MultiSwitchProgramEnvironment& env, CancellableScope&
// PokemonLA::ButtonDetector detector(logger, PokemonLA::ButtonType::ButtonA,);


#if 0
ItemPrinterMaterialDetector detector(COLOR_RED, LANGUAGE);
detector.make_overlays(overlays);
// cout << (int)detector.find_happiny_dust_row_index(env.inference_dispatcher(), console, context) << endl;
// cout << (int)detector.detect_material_quantity(env.inference_dispatcher(), console, context, 2) << endl;

// test OCR for number 1 -> 999. for black text on light background.
// increasing quantity of materials to sell.
for (int i = 1; i < 1000; i++){
context.wait_for_all_requests();
if (i != (int)detector.detect_material_quantity(env.inference_dispatcher(), console, context, 2)){
throw OperationFailedException(
ErrorReport::SEND_ERROR_REPORT,
console,
"OCR didn't match expected value."
);
}
pbf_press_dpad(context, DPAD_UP, 20, 30);
}

// test OCR for number 1 -> 999. for white text on dark background
// decreasing quantity of current materials by selling.
// for (int i = 999; i > 0; i--){
// context.wait_for_all_requests();
// if (i != (int)detector.detect_material_quantity(env.inference_dispatcher(), console, context, 2)){
// throw OperationFailedException(
// ErrorReport::SEND_ERROR_REPORT,
// console,
// "OCR didn't match expected value."
// );
// }
// pbf_press_button(context, BUTTON_A, 30, 150);
// pbf_press_button(context, BUTTON_A, 30, 150);
// pbf_press_button(context, BUTTON_A, 30, 150);
// pbf_press_button(context, BUTTON_A, 30, 150);
// }


#endif

#if 0
ImageRGB32 image("screenshot-20240701-165012250266.png");

BattleBallReader reader(console, Language::English);
cout << reader.read_quantity(image) << endl;

#endif

#if 0
ImageRGB32 image("screenshot-20240630-183016042676.png");
Expand Down
114 changes: 25 additions & 89 deletions ...Programs/Source/PokemonSV/Inference/ItemPrinter/PokemonSV_ItemPrinterMaterialDetector.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -11,6 +11,7 @@
#include "CommonFramework/Exceptions/OperationFailedException.h"
#include "CommonFramework/ImageTypes/ImageViewRGB32.h"
#include "CommonFramework/ImageTypes/ImageRGB32.h"
#include "CommonFramework/ImageTools/ImageStats.h"
#include "CommonFramework/ImageTools/ImageFilter.h"
#include "CommonFramework/OCR/OCR_NumberReader.h"
#include "CommonFramework/Tools/ConsoleHandle.h"
Expand Down Expand Up @@ -84,101 +85,31 @@ int16_t ItemPrinterMaterialDetector::read_number(
Logger& logger, AsyncDispatcher& dispatcher,
const ImageViewRGB32& screen, const ImageFloatBox& box
) const{
int16_t number_result_white_text = read_number_black_or_white_text(logger, dispatcher, screen, box, true);
if (number_result_white_text == -1){
// try looking for black text
return read_number_black_or_white_text(logger, dispatcher, screen, box, false);
}else{
return number_result_white_text;
}

}

int16_t ItemPrinterMaterialDetector::read_number_black_or_white_text(
Logger& logger, AsyncDispatcher& dispatcher,
const ImageViewRGB32& screen, const ImageFloatBox& box,
bool is_white_text
) const{
ImageViewRGB32 cropped = extract_box_reference(screen, box);

std::vector<BlackWhiteRgb32Range> filters;
if (is_white_text){
filters =
{
// white text filters
{0xff808080, 0xffffffff, true},
{0xff909090, 0xffffffff, true},
{0xffa0a0a0, 0xffffffff, true},
{0xffb0b0b0, 0xffffffff, true},
{0xffc0c0c0, 0xffffffff, true},
{0xffd0d0d0, 0xffffffff, true},
{0xffe0e0e0, 0xffffffff, true},
{0xfff0f0f0, 0xffffffff, true},
};
}else{
filters =
{
// black text filters
// {0xff000000, 0xff101010, true},
// {0xff000000, 0xff202020, true},
{0xff000000, 0xff303030, true},
{0xff000000, 0xff404040, true},
{0xff000000, 0xff505050, true},
{0xff000000, 0xff606060, true},
{0xff000000, 0xff707070, true},
{0xff000000, 0xff808080, true},
};
}

std::vector<std::pair<ImageRGB32, size_t>> filtered = to_blackwhite_rgb32_range(
ImageRGB32 filtered = to_blackwhite_rgb32_range(
cropped,
filters
0xff000000, 0xff808080,
true
);
// filtered.save("DebugDumps/test-one-filter-1.png");
bool is_dark_text_light_background = image_stats(filtered).average.sum() > 500;
Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I'm not sure if this is the best way to check if the text is dark vs light. Happy to hear any suggestions.

Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

500 feels too high. IIRC, I've had issues detecting whites in the past with 500 on my "bad" capture card.

What are the actual values for dark vs. light? Maybe pick the mid-point?

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Dark text light background is usually >600. Light text dark background is usually <200.
I've changed it to 400.
Also, I do apply a binary filter first, so that should hopefully help as well.

// std::cout << "Average sum of filtered: "<< std::to_string(image_stats(filtered).average.sum()) << std::endl;

SpinLock lock;
std::map<int16_t, int8_t> candidates;
std::vector<std::unique_ptr<AsyncTask>> tasks(filtered.size());
// for (size_t c = 0; c < filtered.size(); c++){
// filtered[c].first.save("DebugDumps/test-" + std::to_string(c) + ".png");
// }
for (size_t c = 0; c < filtered.size(); c++){
tasks[c] = dispatcher.dispatch([&, c]{
int num = OCR::read_number(logger, filtered[c].first);
WriteSpinLock lg(lock);
candidates[(int16_t)num]++;
});
}

// Wait for everything.
for (auto& task : tasks){
task->wait_and_rethrow_exceptions();
}

std::pair<int16_t, int8_t> best;
std::pair<int16_t, int8_t> second_best;
for (const auto& item : candidates){
logger.log("Candidate OCR: " + std::to_string(item.first) + "; x" + std::to_string(item.second));
if (item.second >= best.second){
second_best = best;
best = item;
}
// std::cout << "Best: " << std::to_string(best.first) << "; x" << std::to_string(best.second) << std::endl;
// std::cout << "Second Best: " << std::to_string(second_best.first) << "; x" << std::to_string(second_best.second) << std::endl;
}

if (best.second > second_best.second + 3){
return best.first;
int16_t number;
if (is_dark_text_light_background){
number = (int16_t)OCR::read_number_waterfill(logger, cropped, 0xff000000, 0xff808080);
}else{
logger.log("Ambiguous or multiple results with normal number OCR. Use Waterfill number OCR.");
if(is_white_text){
return (int16_t)OCR::read_number_waterfill(logger, cropped, 0xff808080, 0xffffffff);
}else{
return (int16_t)OCR::read_number_waterfill(logger, cropped, 0xff000000, 0xff808080);
}
number = (int16_t)OCR::read_number_waterfill(logger, cropped, 0xff808080, 0xffffffff);
}

if (number < 1 || number > 999){
number = -1;
}
return (int16_t)number;
}


// return row number where Happiny dust is located on screen
// keep pressing DPAD_RIGHT until Happiny dust is on screen
// check each row on the screen for Happiny Dust
Expand Down Expand Up @@ -261,11 +192,16 @@ std::vector<int8_t> ItemPrinterMaterialDetector::find_material_value_row_index(
VideoSnapshot snapshot = console.video().snapshot();
int8_t total_rows = 10;
std::vector<int8_t> row_indexes;
SpinLock lock;
std::vector<std::unique_ptr<AsyncTask>> tasks(total_rows);
for (int8_t row_index = 0; row_index < total_rows; row_index++){
int16_t value = read_number(console, dispatcher, snapshot, m_box_mat_value[row_index]);
if (value == material_value){
row_indexes.push_back(row_index);
}
tasks[row_index] = dispatcher.dispatch([&, row_index]{
int16_t value = read_number(console, dispatcher, snapshot, m_box_mat_value[row_index]);
if (value == material_value){
WriteSpinLock lg(lock);
row_indexes.push_back(row_index);
}
});
}

return row_indexes;
Expand Down
6 changes: 0 additions & 6 deletions ...alPrograms/Source/PokemonSV/Inference/ItemPrinter/PokemonSV_ItemPrinterMaterialDetector.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -81,12 +81,6 @@ class ItemPrinterMaterialDetector{
const ImageViewRGB32& screen, const ImageFloatBox& box
) const;

int16_t read_number_black_or_white_text(
Logger& logger, AsyncDispatcher& dispatcher,
const ImageViewRGB32& screen, const ImageFloatBox& box,
bool is_white_text
) const;


private:
Color m_color;
Expand Down
Loading