【Arduino】试验带识别模块的吃鸡助手

在前面的试验中,我们试验了声音触发点击,方面多指操作的辅助功能,

【Arduino】自制声控点击器(吼叫吃鸡助手)-CSDN博客

如果声控模块换成图像识别模块,就是一个自动识别并shot的功能了,

通过这个试验,可以了解arduino串口通信的使用,蓝牙模块和arduino交互就是通过串口交互的方式。

看看arduino自带的串口例子

cpp 复制代码
/*
  Serial Event example

  When new serial data arrives, this sketch adds it to a String.
  When a newline is received, the loop prints the string and clears it.

  A good test for this is to try it with a GPS receiver that sends out
  NMEA 0183 sentences.

  NOTE: The serialEvent() feature is not available on the Leonardo, Micro, or
  other ATmega32U4 based boards.

  created 9 May 2011
  by Tom Igoe

  This example code is in the public domain.

  https://www.arduino.cc/en/Tutorial/BuiltInExamples/SerialEvent
*/

String inputString = "";      // a String to hold incoming data
bool stringComplete = false;  // whether the string is complete

void setup() {
  // initialize serial:
  Serial.begin(9600);
  // reserve 200 bytes for the inputString:
  inputString.reserve(200);
}

void loop() {
  // print the string when a newline arrives:
  if (stringComplete) {
    Serial.println(inputString);
    // clear the string:
    inputString = "";
    stringComplete = false;
  }
}

/*
  SerialEvent occurs whenever a new data comes in the hardware serial RX. This
  routine is run between each time loop() runs, so using delay inside loop can
  delay response. Multiple bytes of data may be available.
*/
void serialEvent() {
  while (Serial.available()) {
    // get the new byte:
    char inChar = (char)Serial.read();
    // add it to the inputString:
    inputString += inChar;
    // if the incoming character is a newline, set a flag so the main loop can
    // do something about it:
    if (inChar == '\n') {
      stringComplete = true;
    }
  }
}

这里我们使用一个yb k210来作为识别设备,将其tx,rx和arduino的tx,rx交叉连接,

在其人脸识别模块的代码里添加串口通信处理serial.send,

python 复制代码
import sensor, image, time, lcd
from maix import KPU
import gc
from modules import ybserial

lcd.init()
sensor.reset()                      # Reset and initialize the sensor. It will
                                    # run automatically, call sensor.run(0) to stop
sensor.set_pixformat(sensor.RGB565) # Set pixel format to RGB565 (or GRAYSCALE)
sensor.set_framesize(sensor.QVGA)   # Set frame size to QVGA (320x240)
sensor.skip_frames(time = 1000)     # Wait for settings take effect.
clock = time.clock()                # Create a clock object to track the FPS.

serial = ybserial()

anchor = (0.1075, 0.126875, 0.126875, 0.175, 0.1465625, 0.2246875, 0.1953125, 0.25375, 0.2440625, 0.351875, 0.341875, 0.4721875, 0.5078125, 0.6696875, 0.8984375, 1.099687, 2.129062, 2.425937)
kpu = KPU()
kpu.load_kmodel("/sd/KPU/yolo_face_detect/face_detect_320x240.kmodel")
kpu.init_yolo2(anchor, anchor_num=9, img_w=320, img_h=240, net_w=320 , net_h=240 ,layer_w=10 ,layer_h=8, threshold=0.5, nms_value=0.2, classes=1)

lm68_kpu = KPU()
print("ready load model")
lm68_kpu.load_kmodel("/sd/KPU/face_detect_with_68landmark/landmark68.kmodel")

RATIO = 0.08
def extend_box(x, y, w, h, scale):
    x1_t = x - scale*w
    x2_t = x + w + scale*w
    y1_t = y - scale*h
    y2_t = y + h + scale*h
    x1 = int(x1_t) if x1_t>1 else 1
    x2 = int(x2_t) if x2_t<320 else 319
    y1 = int(y1_t) if y1_t>1 else 1
    y2 = int(y2_t) if y2_t<240 else 239
    cut_img_w = x2-x1+1
    cut_img_h = y2-y1+1
    return x1, y1, cut_img_w, cut_img_h

while 1:
    gc.collect()
    #print("mem free:",gc.mem_free())
    clock.tick()                    # Update the FPS clock.
    img = sensor.snapshot()
    kpu.run_with_output(img)
    dect = kpu.regionlayer_yolo2()
    fps = clock.fps()
    if len(dect) > 0:
        print("dect:",dect)
        text = 'Hello11\n'
        num = serial.send(text)
        print("num:", num)
        for l in dect :
            x1, y1, cut_img_w, cut_img_h = extend_box(l[0], l[1], l[2], l[3], scale=RATIO) # 扩大人脸框
            face_cut = img.cut(x1, y1, cut_img_w, cut_img_h)
            a = img.draw_rectangle(l[0],l[1],l[2],l[3], color=(0, 255, 0))
            face_cut_128 = face_cut.resize(128, 128)
            face_cut_128.pix_to_ai()
            out = lm68_kpu.run_with_output(face_cut_128, getlist=True)
            #print("out:",len(out))
            for j in range(68):
                x = int(KPU.sigmoid(out[2 * j])*cut_img_w + x1)
                y = int(KPU.sigmoid(out[2 * j + 1])*cut_img_h + y1)
                #a = img.draw_cross(x, y, size=1, color=(0, 0, 255))
                a = img.draw_circle(x, y, 2, color=(0, 0, 255), fill=True)
            del (face_cut_128)
            del (face_cut)

    a = img.draw_string(0, 0, "%2.1ffps" %(fps), color=(0, 60, 255), scale=2.0)
    lcd.display(img)
    gc.collect()

kpu.deinit()
lm68_kpu.deinit()

arduino收到串口信号时,调用继电器来实现通电,达到点击屏幕的效果,

在串口接收上添加了一点修改

cpp 复制代码
/*
  Serial Event example

  When new serial data arrives, this sketch adds it to a String.
  When a newline is received, the loop prints the string and clears it.

  A good test for this is to try it with a GPS receiver that sends out
  NMEA 0183 sentences.

  NOTE: The serialEvent() feature is not available on the Leonardo, Micro, or
  other ATmega32U4 based boards.

  created 9 May 2011
  by Tom Igoe

  This example code is in the public domain.

  https://www.arduino.cc/en/Tutorial/BuiltInExamples/SerialEvent
*/
int relayPin = 4;

String inputString = "";      // a String to hold incoming data
bool stringComplete = false;  // whether the string is complete

void setup() {
  // initialize serial:
  Serial.begin(115200);
  // reserve 200 bytes for the inputString:
  inputString.reserve(200);
      pinMode(relayPin, OUTPUT);
}

void loop() {
  // print the string when a newline arrives:
  if (stringComplete) {
    Serial.println(inputString);
    // clear the string:
    inputString = "";
    
    digitalWrite(relayPin, HIGH);
    delay(3000);
    digitalWrite(relayPin, LOW);
    stringComplete = false;
  }
  delay(10);
}

/*
  SerialEvent occurs whenever a new data comes in the hardware serial RX. This
  routine is run between each time loop() runs, so using delay inside loop can
  delay response. Multiple bytes of data may be available.
*/
void serialEvent() {
  if (stringComplete) {
    return;
  }
  while (Serial.available()) {
    // get the new byte:
    char inChar = (char)Serial.read();
    // add it to the inputString:
    inputString += inChar;
    // if the incoming character is a newline, set a flag so the main loop can
    // do something about it:
    if (inChar == '\n') {
      stringComplete = true;
      break;
    }
  }
}

这里使用的是人脸检测,在吃鸡实战中效果不好,

人体检测会好一些,

去训练的话可以参考

http://t.csdnimg.cn/ZfoSE

【手把手在K210上部署自己在线训练的YOLO模型 - CSDN App】http://t.csdnimg.cn/uwcb9

http://t.csdnimg.cn/uwcb9

相关推荐
神马行空1 小时前
一文解读DeepSeek大模型在政府工作中具体的场景应用
人工智能·大模型·数字化转型·deepseek·政务应用
合合技术团队1 小时前
实测对比|法国 AI 独角兽公司发布的“最强 OCR”,实测效果如何?
大数据·人工智能·图像识别
蒹葭苍苍8731 小时前
LoRA、QLoRA微调与Lama Factory
人工智能·笔记
蹦蹦跳跳真可爱5891 小时前
Python----机器学习(基于PyTorch的线性回归)
人工智能·pytorch·python·机器学习·线性回归
mosquito_lover12 小时前
矿山边坡监测预警系统设计
人工智能·python·深度学习·神经网络·视觉检测
契合qht53_shine2 小时前
OpenCV 从入门到精通(day_03)
人工智能·opencv·计算机视觉
ZaaaaacK2 小时前
STM32技能综合巩固
stm32·单片机·嵌入式硬件
Naomi5212 小时前
Trustworthy Machine Learning
人工智能·机器学习
刘 怼怼3 小时前
使用 Vue 重构 RAGFlow 实现聊天功能
前端·vue.js·人工智能·重构
程序员安仔3 小时前
每天学新 AI 工具好累?我终于发现了“一键全能且免费不限量”的国产终极解决方案
人工智能