(SKU:SEN0315) Gravity：PAJ7620U2 Gesture Sensor 手势识别传感器

产品名称(300px)

简介

PAJ7620U2手势识别传感器是一款以IIC通信的3D手势识别交互式传感器。工作电流仅3.5mA, 手势识别距离达20mm。在最远20cm范围内，可以识别的手势多达13种。该手势传感器具备良好的手势识别稳定性，反应快，准确率高，可识别多种手势。支持两种手势识别模式，高速模式（可识别9种）和低速模式（可识别13种）。支持用户自定义识别手势。

PAJ7620U2手势识别传感器在高速模式下可识别上，下，左，右，前，后，顺时针，逆时针，快速挥手。而低速模式除了支持高速基础包含的所有手势外，还支持慢速上下，慢速左右，慢速前后，乱序。此外，低速模式可以自定义单位采样时间，还可根据自己需要轻松定义识别手势。

市面上主流的手势传感器可识别手势通常仅为六种，而对于非接触式遥控器，手势游戏机等应用，仅有的基础性6种识别模式完全不够。而PAJ7620U2手势识别传感器的出现，解决了手势识别模式欠缺的问题。PAJ7620U2手势识别传感器广泛适用于非接触式遥控器，机器人交互，手势操作游戏机，手势控制酷炫灯光等应用。设想一下，挥挥手控制电视空调；挥挥手调整灯光和音乐；挥挥手去操纵游戏...会不会很有意思呢？

技术规格

PAJ7620U2 Gesture Sensor 手势识别传感器尺寸图

工作电压：3.3~5.5V
工作电流：3.5mA
通信接口：Gravity-IIC 4Pin
IIC地址：0x73
尺寸：30mmx22mm
安装孔尺寸：15mm
可识别距离：30mm~200mm
手势更新率：120Hz
工作温度：-40℃~85℃
环境光免疫力：<100k Lux
盖板材料：推荐采用玻璃或PC;透明度须>90%;盖板材料厚度<0.7mm;盖板和模块应尽可能靠近，最远距离为0.2mm

引脚说明

顶部示意图
底部示意图

丝印	功能描述
D	IIC数据SDA
C	IIC时钟SCL
-	电源负极
+	电源正极

使用教程

准备

硬件
- UNO x1
- IO 传感器拓展板 x1
- PAJ7620U2手势识别传感器x1
软件
- Arduino IDE 点击下载Arduino IDE

接线图

准备好硬件后，按照下图将模块与UNO连接好

样例代码

点击下载库文件例程和库文件。如何安装库？

函数参考：

 DFRobot_PAJ7620U2(TwoWire *pWire=&Wire); 
 /**
  * @brief 构造函数
  * @param mode 构造设备时，可以指定它的默认工作模式
  */

 int begin(void) 
  /**
  * @brief 初始化函数
  * @return 返回0表示初始化成功，返回其他值表示初始化失败
  */

 void setGestureHighRate(bool b);
  /**
  * @brief 设置高速手势识别模式
  * @param b true表示配置为高速识别模式，以最快速度识别手势并返回。
  * @n  false表示低速模式，在低速模式下，系统会做更多的判断
  * @n  在高速识别模式下，可以快速识别的动作包括向左滑动 向右滑动 向上滑动 向下滑动 
  * @n  向前滑动 向后滑动 逆时针 顺时针 快速挥手 9个动作
  * @n  高级用户如果想要用这些动作的组合，需要在外部自己完成算法逻辑的编写，比如左右左快速挥手
  * @n  因为每个人用到的动作手势有限 ，我们没有将更多的扩展动作手势集在库中，需要用户在ino文件中自己完成编程
  * @n
  * @n
  * @n  在低速识别模式下，每2秒识别一个动作，我们将一些扩展动作集成到库内部，方便基础用户使用
  * @n  可以识别的动作包括向左滑动 向右滑动 向上滑动 向下滑动 向前滑动 向后滑动 
  * @n  逆时针 顺时针 快速挥手 9个基础动作 左右慢挥手 上下慢挥手 前后慢挥手 乱序慢挥手  4个扩展动作 
  */

 String gestureDescription(eGesture_t gesture);
  /**
  * @brief 获取手势号码对应的字符串描述
  * @param gesture 包含在eGesture_t中的手势号码
  * @return 手势号码对应的文字描述信息，如果输入了手势表中不存在的手势，返回空字符串
  * @n 正常的返回值可能是   "None","Right","Left", "Up", "Down", "Forward", "Backward", "Clockwise",
  * @n "Anti-Clockwise", "Wave", "WaveSlowlyDisorder", "WaveSlowlyLeftRight", "WaveSlowlyUpDown",
  * @n "WaveSlowlyForwardBackward"
  */

 eGesture_t getGesture(void);
 /**
  * @brief 获取手势
  * @return 返回当前手势，可能是的值为eGestureNone  eGestureRight  eGestureLeft  eGestureUp  
  * @n     eGestureDown  eGestureForward  eGestureBackward  eGestureClockwise
  * @n     eGestureWave  eGestureWaveSlowlyDisorder  eGestureWaveSlowlyLeftRight  
  * @n     eGestureWaveSlowlyUpDown  eGestureWaveSlowlyForwardBackward
  */

高速模式：

代码功能：可以识别以下手势。

注：顺时针和逆时针手势需要至少转两圈以上更多操作详见问答部分参考视频

高速模式可识别手势
gesture code	gesture dsctiption
1	Right
2	Left
4	Up
8	Down
16	Forward
32	Backward
64	Clockwise
128	Anti-clockwise
256	Wave(quickly)
the totality of recognizable gesture is 9

/*!
 * @file GestureRecognize_HighRate.ino
 * @brief Present the 9 built-in gestures data the sensor supports. 
 * @n Wave your hand above the sensor (within 0~20cm), it can recognize 9 kinds of gestures: move up, down, left, right, forward,
 * @n backward, clockwise, anti-clockwise, wave.
 * @n For more usages of the sensor, refer to the description about setGestureHighRate in function setup.
 *
 * @copyright   Copyright (c) 2010 DFRobot Co.Ltd (http://www.dfrobot.com)
 * @licence     The MIT License (MIT)
 * @author      Alexander(ouki.wang@dfrobot.com)
 * @version  V1.0
 * @date  2019-07-16
 * @get from https://www.dfrobot.com
 * @url https://github.com/DFRobot/DFRobot_PAJ7620U2
 */
 
#include <DFRobot_PAJ7620U2.h>

DFRobot_PAJ7620U2 paj;

void setup()
{
  Serial.begin(115200);
  delay(300);
  Serial.println("Gesture recognition system base on PAJ7620U2");
  while(paj.begin() != 0){
    Serial.println("initial PAJ7620U2 failure! Please check if all the connections are fine, or if the wire sequence is correct?");
    delay(500);
  }
  Serial.println("PAJ7620U2 init completed, start to test the gesture recognition function");
  
  /*Set fast detection mode 
   *If the parameter is set to false, the module enters slow detection mode, and it detects one gesture every 2s. We have integrated
   *some gestures inside the module to make it convenient for beginners.
   *The slow mode can recognize 9  basic gestures and 4 expanded gestures: move left, right, up, down, forward, backward, clockwise, 
   *counter-clockwise, wave, slowly move left and right, slowly move up and down, slowly move forward and backward, 
   *wave slowly and randomly.
   *
   *
   *
   *If the parameter is set to true, the module enters fast detection mode. 
   *The fast mode can recognize 9 gestures: move left, right, up, down, forward, backward, clockwise, counter-clockwise, wave
   *To detect the combination of these gestures, like wave left, right and left quickly, users needs to design their own algorithms logic.
   *Since users only use limited gestures in this mode, we are not going to integrate too much expanded gestures in the library. 
   *If necessary, you can complete the algorithm logic in the ino file by yourself.
   */
  paj.setGestureHighRate(true);

}

void loop()
{
  /* Read gesture number（return eGesture_t enumerated type）
   * eGestureNone  eGestureRight  eGestureLeft  eGestureUp  eGestureDown  eGestureForward
   * eGestureBackward  eGestureClockwise  eGestureAntiClockwise  eGestureWave  eGestureWaveSlowlyDisorder
   * eGestureWaveSlowlyLeftRight  eGestureWaveSlowlyUpDown  eGestureWaveSlowlyForwardBackward
   */
  DFRobot_PAJ7620U2::eGesture_t gesture = paj.getGesture();
  if(gesture != paj.eGestureNone ){
   /* Get the string descritpion corresponding to the gesture number.
    * The string description could be 
    * "None","Right","Left", "Up", "Down", "Forward", "Backward", "Clockwise", "Anti-Clockwise", "Wave",
    * "WaveSlowlyDisorder", "WaveSlowlyLeftRight", "WaveSlowlyUpDown", "WaveSlowlyForwardBackward"
    */
    String description  = paj.gestureDescription(gesture);//Convert gesture number into string description
    Serial.println("--------------Gesture Recognition System---------------------------");
    Serial.print("gesture code        = ");Serial.println(gesture);
    Serial.print("gesture description  = ");Serial.println(description);
    Serial.println();
  }
}

结果

结果：可识别9种手势，可以在串口打印出相应的手势描述，如下图所示：

低速模式：

注：低速模式下默认采样时间为2s,所以完成一项操作后可能需等待1s 更多操作详见问答部分参考视频

代码功能：可以识别以下手势。

低速模式可识别手势
gesture code	gesture dsctiption
1	Right
2	Left
3	WaveSlowlyLeftRight
4	Up
8	Down
12	WaveSlowlyUpDown
16	Forward
32	Backward
48	WaveSlowlyForwardBackward
64	Clockwise
128	Anti-clockwise
256	Wave(quickly)
512	WaveSlowlyDisorder
the totality of recognizable gesture is 13

#include <DFRobot_PAJ7620U2.h>
 
/*!
 * @file GestureRecognize_LowRate.ino
 * @brief Present the 9 built-in gestures the sensor supports and 4 extended gestures in the slow mode.  
 * @n Wave you hand above the sensor(within 0~20cm), it can detect: move left, right, up, down, forward, backward, clockwise, 
 * @n anti-clockwise, wave, slowly move left and right, slowly move up and down, slowly move forward and backward, wave randomly and slowly.
 * @n For more usages of the sensor, refer to the description about setGestureLowRate in function setup.  
 *
 * @copyright   Copyright (c) 2010 DFRobot Co.Ltd (http://www.dfrobot.com)
 * @licence     The MIT License (MIT)
 * @author      Alexander(ouki.wang@dfrobot.com)
 * @version  V1.0
 * @date  2019-07-16
 * @get from https://www.dfrobot.com
 * @url https://github.com/DFRobot/DFRobot_PAJ7620U2
 */
 
#include <DFRobot_PAJ7620U2.h>

DFRobot_PAJ7620U2 paj;

void setup()
{
  Serial.begin(115200);
  delay(300);
  Serial.println("Gesture recognition system base on PAJ7620U2");
  while(paj.begin() != 0){
    Serial.println("initial PAJ7620U2 failure! Please check if all the connections are fine, or if the wire sequence is correct?");
    delay(500);
  }
  Serial.println("PAJ7620U2init completed, start to test the gesture recognition function");
  
  /*Set fast detection mode 
   *If the parameter is set to false, the module enters slow detection mode, and it detects one gesture every 2s. We have integrated 
   *some gestures inside the module to make it convenient for beginners.
   *The slow mode can recognize 9  basic gestures and 4 expanded gestures: move left, right, up, down, forward, backward, clockwise, 
   *counter-clockwise, wave, slowly move left and right, slowly move up and down, slowly move forward and backward,
   *wave slowly and randomly. 
   *
   *
   *
   *If the parameter is set to true, the module enters fast detection mode. 
   *The fast mode can recognize 9 gestures: move left, right, up, down, forward, backward, clockwise, counter-clockwise, wave.
   *To detect the combination of these gestures, like wave left, right and left quickly, users needs to design their own
   *algorithms logic.
   *Since users only use limited gestures in this mode, we are not going to integrate too much expanded gestures in the library.
   *If necessary, you can complete the algorithm logic in the ino file by yourself.
   */
  paj.setGestureHighRate(false);

}

void loop()
{
  /* Read gesture number（return eGesture_t enumerated type）
   * eGestureNone  eGestureRight  eGestureLeft  eGestureUp  eGestureDown  eGestureForward
   * eGestureBackward  eGestureClockwise  eGestureAntiClockwise  eGestureWave  eGestureWaveSlowlyDisorder
   * eGestureWaveSlowlyLeftRight  eGestureWaveSlowlyUpDown  eGestureWaveSlowlyForwardBackward
   */
  DFRobot_PAJ7620U2::eGesture_t gesture = paj.getGesture();
  if(gesture != paj.eGestureNone ){
   /* Get the string descritpion corresponding to the gesture number.
    * The string description could be 
    * "None","Right","Left", "Up", "Down", "Forward", "Backward", "Clockwise", "Anti-Clockwise", "Wave",
    * "WaveSlowlyDisorder", "WaveSlowlyLeftRight", "WaveSlowlyUpDown", "WaveSlowlyForwardBackward"
    */
    String description  = paj.gestureDescription(gesture);//Convert gesture number into string description
    Serial.println("--------------Gesture Recognition System---------------------------");
    Serial.print("gesture code        = ");Serial.println(gesture);
    Serial.print("gesture description  = ");Serial.println(description);
    Serial.println();
  }
}

结果

结果：可识别13手势，可以在串口打印出相应的手势描述，如下图所示：

Mind+(基于Scratch3.0)图形化编程

1、下载及安装软件，版本不低于1.6.2 RC2.0。下载地址：http://www.mindplus.cc 详细教程：Mind+基础wiki教程-软件下载安装
2、切换到“上传模式”。详细教程：Mind+基础wiki教程-上传模式编程流程
3、“扩展”中选择“主控板”中的“Arduino Uno”,“用户库”中搜索加载手势。详细教程：Mind+基础wiki教程-加载扩展库流程
4、进行编程，程序如下图。
5、菜单“连接设备”，“上传到设备”
6、程序上传完毕后，打开串口即可看到数据输出。详细教程：Mind+基础wiki教程-串口打印

常见问题

问：有高速模式演示视频吗？

答：https://www.bilibili.com/video/av76484181

问：有低速模式演示视频吗？

答：https://www.bilibili.com/video/av76484092

更多问题及有趣的应用，可以 访问论坛 进行查阅或发帖！