How do I transfer an image taken by ESP32-Cam over sockets? - c++

Question

I have an ESP32-Cam and I would like to save an image that was taken by it to my desktop. The ultimate goal would be to display the image in an Android app, but thats a different story.
In order to do this I thought I have to transfer the image with sockets.
The code beneath shows my example code. I have a very simple python program that connects to the WifiServer and prints whatever client.write() is writing.
However I am not sure how I am supposed to transfer the image in this way. Is this even possible? Because I have to transfer the images buffer (in this case fb->buf) right?
But how would I make an image out of that once I received the buffer e.g. in my python code?
It would be great, if someone could help me to get this running, since I never worked with this (images / sockets) before.
ESP Code
#include "esp_camera.h"
#include "Arduino.h"
#include "WiFi.h"
// Pin definition for CAMERA_MODEL_AI_THINKER
#define PWDN_GPIO_NUM 32
#define RESET_GPIO_NUM -1
#define XCLK_GPIO_NUM 0
#define SIOD_GPIO_NUM 26
#define SIOC_GPIO_NUM 27
#define Y9_GPIO_NUM 35
#define Y8_GPIO_NUM 34
#define Y7_GPIO_NUM 39
#define Y6_GPIO_NUM 36
#define Y5_GPIO_NUM 21
#define Y4_GPIO_NUM 19
#define Y3_GPIO_NUM 18
#define Y2_GPIO_NUM 5
#define VSYNC_GPIO_NUM 25
#define HREF_GPIO_NUM 23
#define PCLK_GPIO_NUM 22
// Variables
camera_fb_t * fb = NULL;
WiFiServer wifiServer(5005);
const char* ssid = "ssid";
const char* password = "password";
void setup() {
Serial.begin(115200);
WiFi.begin(ssid, password);
Serial.print("Connecting to WiFi.");
while (WiFi.status() != WL_CONNECTED) {
delay(1000);
Serial.print(".");
}
Serial.println("\nConnected!");
Serial.println(WiFi.localIP());
wifiServer.begin();
camera_config_t config;
config.ledc_channel = LEDC_CHANNEL_0;
config.ledc_timer = LEDC_TIMER_0;
config.pin_d0 = Y2_GPIO_NUM;
config.pin_d1 = Y3_GPIO_NUM;
config.pin_d2 = Y4_GPIO_NUM;
config.pin_d3 = Y5_GPIO_NUM;
config.pin_d4 = Y6_GPIO_NUM;
config.pin_d5 = Y7_GPIO_NUM;
config.pin_d6 = Y8_GPIO_NUM;
config.pin_d7 = Y9_GPIO_NUM;
config.pin_xclk = XCLK_GPIO_NUM;
config.pin_pclk = PCLK_GPIO_NUM;
config.pin_vsync = VSYNC_GPIO_NUM;
config.pin_href = HREF_GPIO_NUM;
config.pin_sscb_sda = SIOD_GPIO_NUM;
config.pin_sscb_scl = SIOC_GPIO_NUM;
config.pin_pwdn = PWDN_GPIO_NUM;
config.pin_reset = RESET_GPIO_NUM;
config.xclk_freq_hz = 20000000;
config.pixel_format = PIXFORMAT_JPEG;
if (psramFound()) {
config.frame_size = FRAMESIZE_UXGA;
config.jpeg_quality = 10;
config.fb_count = 2;
} else {
config.frame_size = FRAMESIZE_SVGA;
config.jpeg_quality = 12;
config.fb_count = 1;
}
// Init Camera
esp_err_t err = esp_camera_init(&config);
if (err != ESP_OK) {
Serial.printf("Camera init failed with error 0x%x", err);
return;
}
}
void loop() {
WiFiClient client = wifiServer.available();
if (client) {
fb = esp_camera_fb_get();
if(!fb) {
Serial.println("Camera capture failed");
return;
} else {
Serial.println("Camera capture succesfull!");
}
while (client.connected()) {
// How do I transfer the image? What do I have to do?
client.write(fb->buf);
}
client.stop();
Serial.println("Client disconnected");
}
}
Python code
import socket
sock = socket.socket()
host = "192.168.178.103" # esp ip
port = 5005 # esp port
sock.connect((host, port))
data = sock.recv(256)
print(data)
sock.close()

Answer 1

Here is Arduino code that worked for me (starting at the loop function)
//=======================================================================
// Main Program Loop
//=======================================================================
void loop() {
WiFiClient client;
if (!client.connect(host, port)) {
Serial.println("Connection to host failed");
delay(1000);
return;
}
Serial.println("Connected to server successful!");
// capture camera frame
camera_fb_t *fb = esp_camera_fb_get();
if(!fb) {
Serial.println("Camera capture failed");
return;
} else {
Serial.println("Camera capture successful!");
}
const char *data = (const char *)fb->buf;
// Image metadata. Yes it should be cleaned up to use printf if the function is available
Serial.print("Size of image:");
Serial.println(fb->len);
Serial.print("Shape->width:");
Serial.print(fb->width);
Serial.print("height:");
Serial.println(fb->height);
client.print("Shape->width:");
client.print(fb->width);
client.print("height:");
client.println(fb->height);
// Give the server a chance to receive the information before sending an acknowledgement.
delay(1000);
getResponse(client);
Serial.print(data);
client.write(data, fb->len);
esp_camera_fb_return(fb);
Serial.println("Disconnecting...");
client.stop();
delay(2000);
}
void getResponse(WiFiClient client) {
byte buffer[8] = { NULL };
while (client.available() > 0 || buffer[0] == NULL) {
int len = client.available();
Serial.println("Len" + len);
if (len > 8) len = 8;
client.read(buffer, len);
if (printReceivedData) {
Serial.write(buffer, len); // show in the serial monitor (slows some boards)
Serial.println("");
}
}
}
The python code to receive the image
import io
import socket
import time
import numpy as np
from PIL import Image
import connection_and_network_constants
serv = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
# address '0.0.0.0' or '' work to allow connections from other machines. 'localhost' disallows external connections.
# see https://www.raspberrypi.org/forums/viewtopic.php?t=62108
serv.bind(('0.0.0.0', connection_and_network_constants.SOCKET_PORT))
serv.listen(5)
print("Ready to accept 5 connections")
def create_image_from_bytes(image_bytes) -> Image.Image:
stream = io.BytesIO(image_bytes)
return Image.open(stream)
while True:
conn, addr = serv.accept()
array_from_client = bytearray()
shape = None
chunks_received = 0
start = time.time()
shape_string = ''
while True:
# print('waiting for data')
# Try 4096 if unsure what buffer size to use. Large transfer chunk sizes (which require large buffers) can cause corrupted results
data = conn.recv(connection_and_network_constants.BUFFER_SIZE)
if not data or data == b'tx_complete':
break
elif shape is None:
shape_string += data.decode("utf-8")
# Find the end of the line. An index other than -1 will be returned if the end has been found because
# it has been received
if shape_string.find('\r\n') != -1:
width_index = shape_string.find('width:')
height_index = shape_string.find('height:')
width = int(shape_string[width_index + len('width:'): height_index])
height = int(shape_string[height_index + len('height:'): ])
shape = (width, height)
print("shape is {}".format(shape))
else:
chunks_received += 1
# print(chunks_received)
array_from_client.extend(data)
# print(array_from_client)
conn.sendall(b'ack')
# print("sent acknowledgement")
# TODO: need to check if sending acknowledgement of the number of chunks and the total length of the array is a good idea
print("chunks_received {}. Number of bytes {}".format(chunks_received, len(array_from_client)))
img: Image.Image = create_image_from_bytes(array_from_client)
img.show()
array_start_time = time.time()
image_array = np.asarray(img)
print('array conversion took {} s'.format(time.time() - array_start_time))
conn.close()
print('client disconnected')