利用单片机在封闭环境下计算水下物体数量

本文主要是介绍利用单片机在封闭环境下计算水下物体数量，希望对大家解决编程问题提供一定的参考价值，需要的开发者们随着小编来一起学习吧！

提示：文章写完后，目录可以自动生成，如何生成可参考右边的帮助文档

前言

一、架构图及流程图

二、ESP32-CAM代码

三、Web服务器

四、图像处理程序代码

五、前端代码

六、使用流程

总结

前言

随着当今社会的发展，水产品市场作为一个重要的食品供应领域，其规模庞大且竞争激烈。然而，在售卖水产品时，例如螃蟹，虾等，通常需要按照数量来计算和售卖，这个传统的方式不仅耗费了大量的人力和时间资源，还容易引发计算错误，导致盈利能力下降。

传统的水产品销售方式主要依赖于人工计算，这需要雇佣大量的工作人员来记录、计算和核对订单。这不仅增加了人力成本，还容易出现计算错误，导致产品的过多或过少售出，从而损害了盈利能力和客户满意度。此外，传统方式也会导致资源浪费。

所以传统的水产品销售方式在当今社会已经不再适用，因为它浪费了资源，容易出现错误，降低了盈利能力。本文便设计了一种基于单片机的设备，旨在在封闭环境下计算水下物体的数量，从而提高售卖过程的效率和精度。通过采用现代技术和数字化销售方式，水产品市场可以提高效率，减少成本，更好地满足客户和商户的需求，从而在竞争激烈的市场中保持竞争力。

一、架构图及流程图

架构图

流程图

二、ESP32-CAM代码

import machine
import time
import network
import ntptime
import urequests as requests
from machine import Timer, Pin# 声明引脚 13 作为LED的引脚
red_pin = Pin(13, Pin.OUT)
timer = Timer(1)  # 创建定时器对象
(ST_INIT, ST_READY, ST_BUSY) = (300, 0, 100)def toggle_led(red_pin):red_pin.value(not red_pin.value())
#指示灯
def led_blink_timed(timer, red_pin, state):#就绪if state == 'READY':timer.deinit()red_pin.value(ST_READY)#初始elif state == 'INIT':timer.init(period=ST_INIT, mode=Timer.PERIODIC, callback=lambda t: toggle_led(red_pin))#忙碌elif state == 'BUSY':timer.init(period=ST_BUSY, mode=Timer.PERIODIC, callback=lambda t: toggle_led(red_pin))else:print('not define yet')# 定时器触发
led_blink_timed(timer, red_pin, state='INIT')# 连接到 Wi-Fi 网络
def connectWiFi():wlan = network.WLAN(network.STA_IF)wlan.active(True)if not wlan.isconnected():print("连接到Wi-Fi网络...")wlan.connect('ESP32','123456789')while not wlan.isconnected():passprint('网络配置: ', wlan.ifconfig())led_blink_timed(timer, red_pin, state='INIT')
connectWiFi()# freq设置PWM频率，duty设置初始占空比为1023
led_pin = machine.PWM(machine.Pin(4), freq=1000, duty=1023)def led_control(current_brightness, led_pwm):if current_brightness == 1023:current_brightness = 100elif current_brightness == 100:current_brightness = 50elif current_brightness == 50:current_brightness = 1023else:current_brightness = 1023  # 如果当前亮度无效，将其设置为默认值1023led_pwm.duty(current_brightness)  # 更新LED亮度return current_brightness# 在主循环中初始化 LED 亮度
led_brightness = 1023import camera
# 拍照并上传图片
def image_upload():flask_url = 'http://172.20.10.2:5000/esp32_im'led_blink_timed(timer, red_pin, state='BUSY')# 初始化摄像头while not camera.init(0, format=camera.JPEG, fb_location=camera.PSRAM):time.sleep(1)frame = camera.capture()print("拍摄完成")camera.deinit()led_blink_timed(timer, red_pin, state='READY')r = requests.post(flask_url, headers={'content-type': 'image/jpeg'}, data=frame)return r.json(), 200, {'Content-Type': 'application/json'}# 初始化按钮引脚
button_pin = machine.Pin(12, machine.Pin.IN, machine.Pin.PULL_UP)
if __name__ == '__main__':button_state = button_pin.value()  # 初始化按钮状态last_trigger_time = 0  # 上次触发操作的时间last_button_state = 1  # 初始状态为按键未按下last_click_time = 0double_click_timeout = 3  # 双击的最大时间间隔min_double_click_interval = 1  # 最小双击间隔，单位为秒debounce_delay = 50  # 消抖延迟，毫秒while True:current_time = time.ticks_ms()  # 获取当前时间，以便检查是否达到了消抖延迟。# 检查消抖延迟是否满足if current_time - last_trigger_time >= debounce_delay:new_button_state = button_pin.value()  # 读取按钮引脚的当前状态if new_button_state != button_state:# 按钮状态改变button_state = new_button_stateif button_state == 0:# 根据单击或双击执行不同的操作time_since_last_click = current_time - last_click_timeprint(time_since_last_click)if (time_since_last_click > min_double_click_interval * 1000and time_since_last_click < double_click_timeout * 1000):led_brightness = led_control(led_brightness, led_pin)print("亮度:",led_brightness)else:print("单击操作")print("发送照片")image_upload()print("发送成功")last_click_time = current_timelast_trigger_time = current_time

注释：照明LED是ESP32-CAM自带的灯，引脚为4，按钮接的引脚为12。状态灯red，接的引脚为13。

三、Web服务器

from flask import Flask, request, jsonify,render_template
from flask_cors import cross_origin
import time
import os
from countstone import countstones
app = Flask(__name__)
@app.route("/esp32_im", methods=["POST","GET"])
def process_image():print("保存图片1")imageData = request.get_data(parse_form_data=False)filename = time.strftime("%m%d%H%M%S", time.localtime()) + ".jpg"path = "./images/" + str(filename)file = open(path, "wb")file.write(imageData)countstones()print("图片已处理")return jsonify({'msg': '图片已保存'})@app.route("/send_image", methods=["GET"])
def send_image1():# 获取最新的照片发送到前端显示print("开始发送图片到前端")maxnum = []picturePath = './static/finsh/'for fi in os.listdir(picturePath):if fi.endswith(".jpg"):maxnum.append(int(fi[0:10]))maxPhoto = max(maxnum)# print(max(maxnum))path = "./static/finsh/"  + str(maxPhoto) + ".jpg"#frame = open(path, 'rb')print(path)return render_template('index.html', image_url=path)
if __name__ == "__main__":app.run(host='0.0.0.0',debug=True,)

四、图像处理程序代码

import cv2
import os
import numpy as np
from matplotlib.pyplot import figure, imshow, axis,showdef countstones():maxnum = []picturePath = './images'for fi in os.listdir(picturePath):if fi.endswith(".jpg"):maxnum.append(int(fi[0:10]))maxPhoto = max(maxnum)path = "./images/"  + str(maxPhoto) + ".jpg"img = cv2.imread(path)# Apply Gamma=0.4 on the normalised image and then multiply by scaling constant (For 8 bit, c=2 5)gamma_point_four = np.array(255*(img/255)**0.4,dtype='uint8')# Similarly, Apply Gamma=0.8gamma_point_eight = np.array(255*(img/255)**0.9,dtype='uint8')# Display the images in subplotsimg3 = cv2.hconcat([gamma_point_four,gamma_point_eight])frame = cv2.normalize(gamma_point_eight, None, alpha=0, beta=1, norm_type=cv2.NORM_MINMAX, dtype=cv2.CV_32F)# scale to uint8frame = (255*frame).astype(np.uint8)image = frame[80:450, 130:520] # 找出ROIhsv = cv2.cvtColor(image, cv2.COLOR_BGR2HSV)# 转换颜色空间 BGR 到 HSV# 定义HSV中蓝色的范围color_dist = {'red': {'Lower': np.array([0, 43,46]), 'Upper': np.array([13, 255, 255])},'blue': {'Lower': np.array([75, 28, 65]), 'Upper': np.array([120, 255, 220])},'green': {'Lower': np.array([35, 43, 35]), 'Upper': np.array([90, 255, 255])},'pink': {'Lower': np.array([156, 43, 46]), 'Upper': np.array([180, 255, 255])},'yellow': {'Lower': np.array([26, 43, 46]), 'Upper': np.array([34, 255, 255])},}filterStone = 'blue'lower_blue = color_dist[filterStone]['Lower'] upper_blue = color_dist[filterStone]['Upper'] # 设置HSV的阈值使得只取蓝色mask = cv2.inRange(hsv, lower_blue, upper_blue) # 过滤蓝色石头# 将掩膜和图像逐像素相加res = cv2.bitwise_and(image,image, mask= mask)res2rgba = cv2.cvtColor(res, cv2.COLOR_BGR2RGBA)# 透过开运算取得蓝色石头所在位置erode_hsv = cv2.erode(mask, None, iterations=1)dilated_hsv = cv2.dilate(erode_hsv, None, iterations=1)# 去除小的轮廓（噪点）min_contour_area = 2 # 轮廓的最小面积，根据需要调整lunkuo_img=image.copy()#获取轮廓_,binary=cv2.threshold(erode_hsv,0,255,cv2.THRESH_BINARY+cv2.THRESH_OTSU)contours, hierarchy = cv2.findContours(binary, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)filtered_contours = [contour for contour in contours if cv2.contourArea(contour) >= min_contour_area]num_filtered_contours = len(filtered_contours)print("过滤后的轮廓数量:", num_filtered_contours)# 画出轮廓image = cv2.drawContours(lunkuo_img, filtered_contours, -1, (0, 255, 0), 1) cv2.putText(image, str(num_filtered_contours), (10, 30),cv2.FONT_HERSHEY_SIMPLEX, 1, (0, 255, 0), 2)BGRimage = image[:, :, ::-1]finshPath = "./static/finsh/" + str(maxPhoto) + ".jpg"cv2.imwrite(finshPath, image)return num_filtered_contours

注释：水下物体是用蓝色石头代替的。

五、前端代码

<!DOCTYPE html>
<html>
<head><title>用户界面</title><style>body {display: flex;flex-direction: column;justify-content: center;align-items: center;height: 100vh;margin: 0;}</style><meta http-equiv="refresh" content="5">
</head>
<body><h1>用户界面</h1><img src="{{ image_url }}" alt="Latest Image" width="700" height="700">
</body>
</html>