fix: move rle into utils

Signed-off-by: badai-nguyen <dai.nguyen@tier4.jp>

perception/autoware_tensorrt_yolox/CMakeLists.txt

@@ -143,6 +143,7 @@ rclcpp_components_register_node(yolox_single_image_inference_node
 )
 
 ament_auto_add_library(${PROJECT_NAME}_node SHARED
+  src/utils.cpp
   src/tensorrt_yolox_node.cpp
 )
 

perception/autoware_tensorrt_yolox/include/autoware/tensorrt_yolox/tensorrt_yolox_node.hpp

@@ -71,7 +71,6 @@ class TrtYoloXNode : public rclcpp::Node
 
 public:
   explicit TrtYoloXNode(const rclcpp::NodeOptions & node_options);
-  std::vector<std::pair<uint8_t, int>> rle_compress(const cv::Mat & mask);
 
 private:
   void onConnect();

perception/autoware_tensorrt_yolox/include/autoware/tensorrt_yolox/utils.hpp

@@ -0,0 +1,25 @@
+// Copyright 2024 TIER IV, Inc.
+//
+// Licensed under the Apache License, Version 2.0 (the "License");
+// you may not use this file except in compliance with the License.
+// You may obtain a copy of the License at
+//
+//     http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing, software
+// distributed under the License is distributed on an "AS IS" BASIS,
+// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+// See the License for the specific language governing permissions and
+// limitations under the License.
+
+#ifndef AUTOWARE__TENSORRT_YOLOX__UTILS_HPP_
+#define AUTOWARE__TENSORRT_YOLOX__UTILS_HPP_
+#include <opencv2/opencv.hpp>
+
+namespace autoware::tensorrt_yolox
+{
+std::vector<std::pair<uint8_t, int>> runLengthEncoder(const cv::Mat & mask);
+cv::Mat runLengthDecoder(const std::vector<uint8_t> & rle_data, const int rows, const int cols);
+}  // namespace autoware::tensorrt_yolox
+
+#endif  // AUTOWARE__TENSORRT_YOLOX__UTILS_HPP_

perception/autoware_tensorrt_yolox/src/tensorrt_yolox_node.cpp

@@ -14,6 +14,7 @@
 
 #include "autoware/tensorrt_yolox/tensorrt_yolox_node.hpp"
 
+#include "autoware/tensorrt_yolox/utils.hpp"
 #include "object_recognition_utils/object_classification.hpp"
 
 #include <autoware_perception_msgs/msg/object_classification.hpp>
@@ -126,25 +127,6 @@ void TrtYoloXNode::onConnect()
   }
 }
 
-std::vector<std::pair<uint8_t, int>> TrtYoloXNode::rle_compress(const cv::Mat & image)
-{
-  std::vector<std::pair<uint8_t, int>> compressed_data;
-  const int rows = image.rows;
-  const int cols = image.cols;
-  compressed_data.emplace_back(image.at<uint8_t>(0, 0), 0);
-  for (int i = 0; i < rows; ++i) {
-    for (int j = 0; j < cols; ++j) {
-      uint8_t current_value = image.at<uint8_t>(i, j);
-      if (compressed_data.back().first == current_value) {
-        ++compressed_data.back().second;
-      } else {
-        compressed_data.emplace_back(current_value, 1);
-      }
-    }
-  }
-  return compressed_data;
-}
-
 void TrtYoloXNode::onImage(const sensor_msgs::msg::Image::ConstSharedPtr msg)
 {
   stop_watch_ptr_->toc("processing_time", true);
@@ -196,14 +178,13 @@ void TrtYoloXNode::onImage(const sensor_msgs::msg::Image::ConstSharedPtr msg)
       overlapSegmentByRoi(yolox_object, mask, width, height);
     }
   }
-  // Compress mask to RLE format
   if (trt_yolox_->getMultitaskNum() > 0) {
     sensor_msgs::msg::Image::SharedPtr out_mask_msg =
       cv_bridge::CvImage(std_msgs::msg::Header(), sensor_msgs::image_encodings::MONO8, mask)
         .toImageMsg();
     out_mask_msg->header = msg->header;
 
-    std::vector<std::pair<uint8_t, int>> compressed_data = TrtYoloXNode::rle_compress(mask);
+    std::vector<std::pair<uint8_t, int>> compressed_data = runLengthEncoder(mask);
     int step = sizeof(uint8_t) + sizeof(int);
     out_mask_msg->data.resize(static_cast<int>(compressed_data.size()) * step);
     for (size_t i = 0; i < compressed_data.size(); ++i) {

perception/autoware_tensorrt_yolox/src/utils.cpp

@@ -0,0 +1,64 @@
+// Copyright 2024 Tier IV, Inc.
+//
+// Licensed under the Apache License, Version 2.0 (the "License");
+// you may not use this file except in compliance with the License.
+// You may obtain a copy of the License at
+//
+//     http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing, software
+// distributed under the License is distributed on an "AS IS" BASIS,
+// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+// See the License for the specific language governing permissions and
+// limitations under the License.
+
+#include "autoware/tensorrt_yolox/utils.hpp"
+
+namespace autoware::tensorrt_yolox
+{
+
+std::vector<std::pair<uint8_t, int>> runLengthEncoder(const cv::Mat & image)
+{
+  std::vector<std::pair<uint8_t, int>> compressed_data;
+  const int rows = image.rows;
+  const int cols = image.cols;
+  compressed_data.emplace_back(image.at<uint8_t>(0, 0), 0);
+  for (int i = 0; i < rows; ++i) {
+    for (int j = 0; j < cols; ++j) {
+      uint8_t current_value = image.at<uint8_t>(i, j);
+      if (compressed_data.back().first == current_value) {
+        ++compressed_data.back().second;
+      } else {
+        compressed_data.emplace_back(current_value, 1);
+      }
+    }
+  }
+  return compressed_data;
+}
+
+cv::Mat runLengthDecoder(const std::vector<uint8_t> & rle_data, const int rows, const int cols)
+{
+  cv::Mat mask(rows, cols, CV_8UC1, cv::Scalar(0));
+  int idx = 0;
+  int step = sizeof(uint8_t) + sizeof(int);
+  int nb_pixels = 0;
+  for (size_t i = 0; i < rle_data.size(); i += step) {
+    uint8_t value;
+    int length;
+    std::memcpy(&value, &rle_data[i], sizeof(uint8_t));
+    std::memcpy(&length, &rle_data[i + sizeof(uint8_t)], sizeof(int));
+    nb_pixels += length;
+    for (int j = 0; j < length; ++j) {
+      int row_idx = static_cast<int>(idx / cols);
+      int col_idx = static_cast<int>(idx % cols);
+      mask.at<uint8_t>(row_idx, col_idx) = value;
+      idx++;
+      if (idx > rows * cols) {
+        break;
+      }
+    }
+  }
+  return mask;
+}
+
+}  // namespace autoware::tensorrt_yolox