BodyBalanceEvaluation/backend/tests/azure_kinect_image_example.py

#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
Azure Kinect SDK 1.4.0 图像获取示例

这个示例展示了如何使用pyKinectAzure库来获取Azure Kinect DK传感器的各种图像数据：
- 彩色图像 (Color Image)
- 深度图像 (Depth Image) 
- 红外图像 (Infrared Image)
- 彩色深度图像 (Colored Depth Image)

要求:
- Azure Kinect SDK 1.4.0
- pykinect_azure库
- OpenCV (cv2)
- numpy

使用方法:
1. 确保Azure Kinect DK设备已连接
2. 运行脚本: python azure_kinect_image_example.py
3. 按 'q' 键退出程序
4. 按 '1' 键切换到彩色图像模式
5. 按 '2' 键切换到深度图像模式  
6. 按 '3' 键切换到红外图像模式
7. 按 '4' 键切换到彩色深度图像模式
8. 按 's' 键保存当前图像
"""

import cv2
import numpy as np
import os
from datetime import datetime

try:
    import pykinect_azure as pykinect
except ImportError:
    print("错误: 无法导入pykinect_azure库")
    print("请使用以下命令安装: pip install pykinect_azure")
    exit(1)

class AzureKinectImageCapture:
    """Azure Kinect图像捕获类"""
    
    def __init__(self):
        self.device = None
        self.current_mode = 1  # 1:彩色, 2:深度, 3:红外, 4:彩色深度
        self.save_counter = 0
        
        # 创建保存图像的目录
        self.save_dir = "captured_images"
        if not os.path.exists(self.save_dir):
            os.makedirs(self.save_dir)
    
    def initialize_device(self):
        """初始化Azure Kinect设备"""
        try:
            # 初始化库，解决"Compatible Azure Kinect SDK not found"兼容性问题
            # 
            # 方法1: 自动检测（可能遇到版本兼容性问题）
            # pykinect.initialize_libraries()
            #
            # 方法2: 手动指定路径（推荐，解决版本兼容性问题）
            # 可选的SDK路径（按优先级排序）：
            import platform
            import os
            
            sdk_paths = []
            if platform.system() == "Windows":
                # Orbbec SDK K4A Wrapper（推荐）
                sdk_paths.append(r"D:\OrbbecSDK_K4A_Wrapper_v1.10.3_windows_202408091749\bin\k4a.dll")
                
               
            else:
                # Linux路径
                sdk_paths.extend([
                    "/usr/lib/x86_64-linux-gnu/libk4a.so",
                    "/usr/local/lib/libk4a.so"
                ])
            
            # 尝试使用可用的SDK路径
            sdk_initialized = False
            for sdk_path in sdk_paths:
                if os.path.exists(sdk_path):
                    try:
                        print(f"尝试使用SDK路径: {sdk_path}")
                        pykinect.initialize_libraries(module_k4a_path=sdk_path)
                        print(f"✓ 成功使用SDK: {sdk_path}")
                        sdk_initialized = True
                        break
                    except Exception as e:
                        print(f"✗ SDK路径失败: {sdk_path} - {e}")
                        continue
            
            # 如果所有手动路径都失败，尝试自动检测
            if not sdk_initialized:
                try:
                    print("尝试自动检测SDK...")
                    pykinect.initialize_libraries()
                    print("✓ 自动检测SDK成功")
                    sdk_initialized = True
                except Exception as e:
                    print(f"✗ 自动检测失败: {e}")
            
            if not sdk_initialized:
                raise Exception("无法初始化Azure Kinect SDK。请检查SDK安装或手动指定正确的路径。")
            
            # 配置设备参数
            device_config = pykinect.default_configuration
            device_config.color_resolution = pykinect.K4A_COLOR_RESOLUTION_1080P  # 1080p彩色分辨率
            device_config.depth_mode = pykinect.K4A_DEPTH_MODE_WFOV_2X2BINNED    # 宽视场深度模式
            device_config.camera_fps = pykinect.K4A_FRAMES_PER_SECOND_30          # 30 FPS
            device_config.synchronized_images_only = True                          # 同步图像
            
            print("设备配置:")
            print(device_config)
            
            # 启动设备
            self.device = pykinect.start_device(config=device_config)
            print("Azure Kinect设备初始化成功!")
            return True
            
        except Exception as e:
            print(f"设备初始化失败: {e}")
            print("请确保:")
            print("1. Azure Kinect DK设备已正确连接")
            print("2. Azure Kinect SDK 1.4.0已正确安装")
            print("3. 设备驱动程序已安装")
            return False
    
    def get_color_image(self):
        """获取彩色图像"""
        capture = self.device.update()
        ret, color_image = capture.get_color_image()
        return ret, color_image
    
    def get_depth_image(self):
        """获取原始深度图像"""
        capture = self.device.update()
        ret, depth_image = capture.get_depth_image()
        return ret, depth_image
    
    def get_colored_depth_image(self):
        """获取彩色深度图像"""
        capture = self.device.update()
        ret, colored_depth_image = capture.get_colored_depth_image()
        return ret, colored_depth_image
    
    def get_infrared_image(self):
        """获取红外图像"""
        capture = self.device.update()
        ret, ir_image = capture.get_ir_image()
        return ret, ir_image
    
    def save_image(self, image, mode_name):
        """保存图像到文件"""
        timestamp = datetime.now().strftime("%Y%m%d_%H%M%S")
        filename = f"{mode_name}_{timestamp}_{self.save_counter:03d}.png"
        filepath = os.path.join(self.save_dir, filename)
        
        cv2.imwrite(filepath, image)
        print(f"图像已保存: {filepath}")
        self.save_counter += 1
    
    def display_instructions(self):
        """显示操作说明"""
        instructions = [
            "=== Azure Kinect 图像捕获示例 ===",
            "操作说明:",
            "1 - 切换到彩色图像模式",
            "2 - 切换到深度图像模式", 
            "3 - 切换到红外图像模式",
            "4 - 切换到彩色深度图像模式",
            "s - 保存当前图像",
            "q - 退出程序",
            "================================="
        ]
        
        for instruction in instructions:
            print(instruction)
    
    def run(self):
        """运行主循环"""
        if not self.initialize_device():
            return
        
        self.display_instructions()
        
        # 创建窗口
        cv2.namedWindow('Azure Kinect Image', cv2.WINDOW_NORMAL)
        cv2.resizeWindow('Azure Kinect Image', 1280, 720)
        
        mode_names = {
            1: "Color Image",
            2: "Depth Image", 
            3: "Infrared Image",
            4: "Colored Depth Image"
        }
        
        print(f"\n当前模式: {mode_names[self.current_mode]}")
        
        while True:
            try:
                # 根据当前模式获取相应图像
                if self.current_mode == 1:
                    ret, image = self.get_color_image()
                elif self.current_mode == 2:
                    ret, image = self.get_depth_image()
                    if ret:
                        # 将深度图像转换为可视化格式
                        image = cv2.convertScaleAbs(image, alpha=0.05)
                elif self.current_mode == 3:
                    ret, image = self.get_infrared_image()
                    if ret:
                        # 将红外图像转换为可视化格式
                        image = cv2.convertScaleAbs(image, alpha=0.5)
                elif self.current_mode == 4:
                    ret, image = self.get_colored_depth_image()
                
                if not ret:
                    continue
                
                # 在图像上添加模式信息
                mode_text = f"Mode: {mode_names[self.current_mode]}"
                cv2.putText(image, mode_text, (10, 30), cv2.FONT_HERSHEY_SIMPLEX, 
                           1, (0, 255, 0), 2, cv2.LINE_AA)
                
                # 显示图像
                cv2.imshow('Azure Kinect Image', image)
                
                # 处理按键
                key = cv2.waitKey(1) & 0xFF
                
                if key == ord('q'):
                    print("退出程序...")
                    break
                elif key == ord('1'):
                    self.current_mode = 1
                    print(f"切换到: {mode_names[self.current_mode]}")
                elif key == ord('2'):
                    self.current_mode = 2
                    print(f"切换到: {mode_names[self.current_mode]}")
                elif key == ord('3'):
                    self.current_mode = 3
                    print(f"切换到: {mode_names[self.current_mode]}")
                elif key == ord('4'):
                    self.current_mode = 4
                    print(f"切换到: {mode_names[self.current_mode]}")
                elif key == ord('s'):
                    self.save_image(image, mode_names[self.current_mode].replace(" ", "_"))
                    
            except KeyboardInterrupt:
                print("\n程序被用户中断")
                break
            except Exception as e:
                print(f"运行时错误: {e}")
                break
        
        # 清理资源
        cv2.destroyAllWindows()
        print("程序结束")

def main():
    """主函数"""
    print("Azure Kinect SDK 1.4.0 图像获取示例")
    print("作者: AI Assistant")
    print("版本: 1.0\n")
    
    # 创建并运行图像捕获实例
    capture = AzureKinectImageCapture()
    capture.run()

if __name__ == "__main__":
    main()