Mirko Pavleski
Published © GPL3+

Simple ESP32 CAM Object detection using Open CV

This is certainly the simplest and cheapest object detection system that can serve perfectly for the presentation.

BeginnerFull instructions provided2 hours685
Simple ESP32 CAM Object detection using Open CV

Things used in this project

Hardware components

ESP32 Camera module
×1

Software apps and online services

Arduino IDE
Arduino IDE

Hand tools and fabrication machines

Multitool, Screwdriver
Multitool, Screwdriver

Story

Read more

Code

Arduino code

C/C++
.
#include <WebServer.h>
#include <WiFi.h>
#include <esp32cam.h>

//THIS PROGRAM SENDS IMAGE IF IT IS PLACED IN WEB IP, BUT IF IT IS PLACED IN PYTHON IT SENDS VIDEO THROUGH THE ITERATIONS. . . (IF IT WORKS IN PYTHON)
const char* WIFI_SSID = "ESP Repeater";
const char* WIFI_PASS = "77777777";

WebServer server(80); //server on port 80

static auto loRes = esp32cam::Resolution::find(320, 240); //low resolution
static auto hiRes = esp32cam::Resolution::find(800, 600); //high resolution
//static auto hiRes = esp32cam::Resolution::find(640, 480); //high resolution (for fps rates) (IP CAM APP)

void
serveJpg() //capture image .jpg
{
  auto frame = esp32cam::capture();
  if (frame == nullptr) {
    Serial.println("Capture Fail");
    server.send(503, "", "");
    return;
  }
  Serial.printf("CAPTURE OK %dx%d %db\n", frame->getWidth(), frame->getHeight(),
                static_cast<int>(frame->size()));

  server.setContentLength(frame->size());
  server.send(200, "image/jpeg");
  WiFiClient client = server.client();
  frame->writeTo(client);  //and send to a client (in this case it will be python)
}

void
handleJpgLo()  //allows to send low resolution image
{
  if (!esp32cam::Camera.changeResolution(loRes)) {
    Serial.println("SET-LO-RES FAIL");
  }
  serveJpg();
}

void
handleJpgHi() //allows to send high resolution image
{
  if (!esp32cam::Camera.changeResolution(hiRes)) {
    Serial.println("SET-HI-RES FAIL");
  }
  serveJpg();
}

void setup()
{
  Serial.begin(115200);
  Serial.println();

  {
    using namespace esp32cam;
    Config cfg;
    cfg.setPins(pins::AiThinker);
    cfg.setResolution(hiRes);
    cfg.setBufferCount(2);
    cfg.setJpeg(80);

    bool ok = Camera.begin(cfg);
    Serial.println(ok ? "CAMERA OK" : "CAMERA FAIL");
  }

  WiFi.persistent(false);
  WiFi.mode(WIFI_STA);
  WiFi.begin(WIFI_SSID, WIFI_PASS); //connect to the WiFi network
  while (WiFi.status() != WL_CONNECTED) {
    delay(500);
  }

  Serial.print("http://");
  Serial.print(WiFi.localIP());
  Serial.println("/cam-lo.jpg");//to connect IP low res

  Serial.print("http://");
  Serial.print(WiFi.localIP());
  Serial.println("/cam-hi.jpg");//to connect high res IP
  server.on("/cam-lo.jpg",handleJpgLo);//send to the server
  server.on("/cam-hi.jpg", handleJpgHi);

  server.begin();
}

void loop()
{
  server.handleClient();
}

Python code

Python
.
import cv2 #opencv
import urllib.request #to open and read URL
import numpy as np

#OBJECT CLASSIFICATION PROGRAM FOR VIDEO IN IP ADDRESS

url = 'http://192.168.100.14/cam-hi.jpg'
#url = 'http://192.168.0.159/'
winName = 'ESP32 CAMERA'
cv2.namedWindow(winName,cv2.WINDOW_AUTOSIZE)
#scale_percent = 80 # percent of original size    #for image processing

classNames = []
classFile = 'coco.names'
with open(classFile,'rt') as f:
    classNames = f.read().rstrip('\n').split('\n')

configPath = 'ssd_mobilenet_v3_large_coco_2020_01_14.pbtxt'
weightsPath = 'frozen_inference_graph.pb'

net = cv2.dnn_DetectionModel(weightsPath,configPath)
net.setInputSize(320,320)
#net.setInputSize(480,480)
net.setInputScale(1.0/127.5)
net.setInputMean((127.5, 127.5, 127.5))
net.setInputSwapRB(True)

while(1):
    imgResponse = urllib.request.urlopen (url) # here open the URL
    imgNp = np.array(bytearray(imgResponse.read()),dtype=np.uint8)
    img = cv2.imdecode (imgNp,-1) #decodificamos

    img = cv2.rotate(img, cv2.ROTATE_90_CLOCKWISE) # vertical
    #img = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY) #black and white

    

    classIds, confs, bbox = net.detect(img,confThreshold=0.5)
    print(classIds,bbox)

    if len(classIds) != 0:
        for classId, confidence,box in zip(classIds.flatten(),confs.flatten(),bbox):
            cv2.rectangle(img,box,color=(0,255,0),thickness = 3) #mostramos en rectangulo lo que se encuentra
            cv2.putText(img, classNames[classId-1], (box[0]+10,box[1]+30), cv2.FONT_HERSHEY_COMPLEX, 1, (0,255,0),2)


    cv2.imshow(winName,img) #  show the picture

    #wait for ESC to be pressed to end the program
    tecla = cv2.waitKey(5) & 0xFF
    if tecla == 27:
        break
cv2.destroyAllWindows()

Libraries

Python
.
No preview (download only).

Credits

Mirko Pavleski

Mirko Pavleski

141 projects • 1241 followers

Comments

Related channels and tags
  • Machine Learning & AI
  • Python on Hardware