把去年的二维码检测、图生文客户端、锥桶检测v8给加了进来，二维码检测新增深度相机检测

src/vlm_detect/package.xml

@@ -0,0 +1,18 @@
+<?xml version="1.0"?>
+<?xml-model href="http://download.ros.org/schema/package_format3.xsd" schematypens="http://www.w3.org/2001/XMLSchema"?>
+<package format="3">
+  <name>vlm_detect</name>
+  <version>0.0.0</version>
+  <description>TODO: Package description</description>
+  <maintainer email="root@todo.todo">root</maintainer>
+  <license>TODO: License declaration</license>
+
+  <test_depend>ament_copyright</test_depend>
+  <test_depend>ament_flake8</test_depend>
+  <test_depend>ament_pep257</test_depend>
+  <test_depend>python3-pytest</test_depend>
+
+  <export>
+    <build_type>ament_python</build_type>
+  </export>
+</package>

src/vlm_detect/setup.cfg

@@ -0,0 +1,4 @@
+[develop]
+script_dir=$base/lib/vlm_detect
+[install]
+install_scripts=$base/lib/vlm_detect

src/vlm_detect/setup.py

@@ -0,0 +1,26 @@
+from setuptools import find_packages, setup
+
+package_name = 'vlm_detect'
+
+setup(
+    name=package_name,
+    version='0.0.0',
+    packages=find_packages(exclude=['test']),
+    data_files=[
+        ('share/ament_index/resource_index/packages',
+            ['resource/' + package_name]),
+        ('share/' + package_name, ['package.xml']),
+    ],
+    install_requires=['setuptools'],
+    zip_safe=True,
+    maintainer='root',
+    maintainer_email='root@todo.todo',
+    description='TODO: Package description',
+    license='TODO: License declaration',
+    tests_require=['pytest'],
+    entry_points={
+      'console_scripts': [
+        'vlm_node = vlm_detect.vlm_node:main',
+      ],
+    },
+)

src/vlm_detect/test/test_copyright.py

@@ -0,0 +1,25 @@
+# Copyright 2015 Open Source Robotics Foundation, 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.
+
+from ament_copyright.main import main
+import pytest
+
+
+# Remove the `skip` decorator once the source file(s) have a copyright header
+@pytest.mark.skip(reason='No copyright header has been placed in the generated source file.')
+@pytest.mark.copyright
+@pytest.mark.linter
+def test_copyright():
+    rc = main(argv=['.', 'test'])
+    assert rc == 0, 'Found errors'

src/vlm_detect/test/test_flake8.py

@@ -0,0 +1,25 @@
+# Copyright 2017 Open Source Robotics Foundation, 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.
+
+from ament_flake8.main import main_with_errors
+import pytest
+
+
+@pytest.mark.flake8
+@pytest.mark.linter
+def test_flake8():
+    rc, errors = main_with_errors(argv=[])
+    assert rc == 0, \
+        'Found %d code style errors / warnings:\n' % len(errors) + \
+        '\n'.join(errors)

src/vlm_detect/test/test_pep257.py

@@ -0,0 +1,23 @@
+# Copyright 2015 Open Source Robotics Foundation, 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.
+
+from ament_pep257.main import main
+import pytest
+
+
+@pytest.mark.linter
+@pytest.mark.pep257
+def test_pep257():
+    rc = main(argv=['.', 'test'])
+    assert rc == 0, 'Found code style errors / warnings'

src/vlm_detect/vlm_detect/vlm_node.py

@@ -0,0 +1,144 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+import rclpy
+from rclpy.node import Node
+from std_msgs.msg import Int32, String
+from sensor_msgs.msg import CompressedImage
+from cv_bridge import CvBridge
+import cv2
+import base64
+import threading
+from openai import OpenAI
+import os
+import time
+import numpy as np
+
+class VLMProcessor(Node):
+    def __init__(self):
+        super().__init__('vlm_dtetct')
+        
+        # 初始化 OpenAI 客户端
+        self.client = OpenAI(
+            base_url="http://192.168.1.103:8000/v1",
+            api_key="EMPTY"
+        )
+        
+        # ROS2 组件
+        self.bridge = CvBridge()
+        self.latest_image = None
+        self.image_lock = threading.Lock()
+        
+        # 订阅图像话题
+        self.image_sub = self.create_subscription(
+            CompressedImage,
+            '/image',
+            self.image_callback,
+            10
+        )
+        
+        # 订阅触发信号
+        self.sign_sub = self.create_subscription(
+            Int32,
+            '/sign4return',
+            self.sign_callback,
+            10
+        )
+        
+        # 发布结果
+        self.result_pub = self.create_publisher(
+            String,
+            '/vlm_result',
+            10
+        )
+        
+        self.get_logger().info("VLM Processor...")
+
+    def image_callback(self, msg):
+        """保存最新的图像"""
+        with self.image_lock:
+            try:
+                np_arr = np.frombuffer(msg.data, np.uint8)
+                # 使用 OpenCV 解码图像
+                cv_image = cv2.imdecode(np_arr, cv2.IMREAD_COLOR)
+                self.latest_image = cv_image
+                self.get_logger().debug("recive picture")
+            except Exception as e:
+                self.get_logger().error("err picture")
+
+    def sign_callback(self, msg):
+        """处理触发信号"""
+        if msg.data == 9:
+            self.get_logger().info("收到触发信号 (6)，开始处理图像...")
+            
+            # 检查是否有可用图像
+            with self.image_lock:
+                if self.latest_image is None:
+                    self.get_logger().warning("没有可用图像")
+                    return
+                
+                # 保存临时图像文件
+                temp_path = "/tmp/vlm_temp_image.jpg"
+                cv2.imwrite(temp_path, self.latest_image)
+                self.get_logger().info("已保存临时图像}")
+            
+            # 处理图像
+            try:
+                description = self.process_image(temp_path)
+                self.get_logger().info(f"图像描述结果:")
+                
+                # 发布结果
+                result_msg = String()
+                result_msg.data = description
+                self.result_pub.publish(result_msg)
+                
+                # 清理临时文件
+                os.remove(temp_path)
+            except Exception as e:
+                self.get_logger().error(f"处理图像时出错:{e}")
+
+    def process_image(self, image_path):
+        """使用 VLM 模型处理图像"""
+        # 读取并编码图像
+        with open(image_path, "rb") as image_file:
+            base64_image = base64.b64encode(image_file.read()).decode('utf-8')
+        
+        # 发送请求到 VLM 模型
+        start_time = time.time()
+        
+        response = self.client.chat.completions.create(
+            model="/home/yyh/vllm_test/model/InternVL3-1B",
+            messages=[
+                {
+                    "role": "user",
+                    "content": [
+                        {"type": "text", "text": "描述图片中的动漫病人"},
+                        {
+                            "type": "image_url",
+                            "image_url": {
+                                "url": f"data:image/jpeg;base64,{base64_image}"
+                            }
+                        }
+                    ]
+                }
+            ],
+            max_tokens=100
+        )
+        
+        processing_time = time.time() - start_time
+        self.get_logger().info("VLM 处理耗时秒")
+        
+        return response.choices[0].message.content
+
+def main(args=None):
+    rclpy.init(args=args)
+    node = VLMProcessor()
+    try:
+        rclpy.spin(node)
+    except KeyboardInterrupt:
+        pass
+    finally:
+        node.destroy_node()
+        rclpy.shutdown()
+
+if __name__ == '__main__':
+    main()