// // Copyright 2019 Miral Shah // // Use, modification and distribution are subject to the Boost Software License, // Version 1.0. (See accompanying file LICENSE_1_0.txt or copy at // http://www.boost.org/LICENSE_1_0.txt) // #include #include using namespace boost::gil; // Demonstrates thresholding // Thresholding can either attribute an arbitrary value to pixels whose values are greater than the threshold // or can truncate the pixel values at an arbitrary maximum. // In particular, the function threshold_truncate accepts a mode and a direction. // Passing threshold_truncate_mode::threshold effectively truncates the values to the threshold, whereas // threshold_truncate_mode::zero sets them to 0. // The combination of mode and direction controls which pixels are modified: // - threshold and regular: truncates the pixels whose values are greater than the threshold // - threshold and inverse: truncates the pixels whose values are less than the threshold // - zero and regular: zeroes the pixels whose values are less than the threshold // - zero and inverse: zeroes the pixels whose values are greater than the threshold // See also: // adaptive_threshold.cpp - Adaptive thresholding int main() { rgb8_image_t img; read_image("test.jpg",img, jpeg_tag{}); rgb8_image_t img_out(img.dimensions()); // performing binary threshold on each channel of the image // if the pixel value is more than 150 than it will be set to 255 else to 0 boost::gil::threshold_binary(const_view(img), view(img_out), 150, 255); write_view("out-threshold-binary.jpg", view(img_out), jpeg_tag{}); // if the pixel value is more than 150 than it will be set to 150 else no change boost::gil::threshold_truncate(const_view(img), view(img_out), 150, threshold_truncate_mode::threshold); write_view("out-threshold-binary_inv.jpg", view(img_out), jpeg_tag{}); return 0; }