How to earn money with images by building an API

It is a very popular use case. A quick search tells us that over ninety thousand people search for "remove background from image" on Google every month!

Many websites and tools are providing this service. You can upload your images there and have their backgrounds removed in no time. However, their pricing is jaw-droppingly expensive, with some charging up to $1.99 per image.

But, hold on, we've got a solution that will give them a run for their money. We can create a similar code that removes backgrounds from images. And you'll be amazed at how easy it is. The best part? You'll be able to make some serious cash with that ready-to-use Python script! It's super simple to use, and we'll show you how to create it into an API to sell it on RapidAPI Hub.

Let’s Start

First, download and install Visual Studio Code and Python to get started. Once done, simply paste the following Python script into your code editor, and voila!

python
Copy
import torch
import torch.nn as nn
import torch.optim as optim
import numpy as np
import cv2
import uuid
import os
from model import U2NET
from torch.autograd import Variable
from skimage import io, transform
from PIL import Image
# Get The Current Directory
currentDir = os.path.dirname(__file__)
# Functions:
# Save Results
def save_output(image_name, output_name, pred, d_dir, type):
    predict = pred
    predict = predict.squeeze()
    predict_np = predict.cpu().data.numpy()
    im = Image.fromarray(predict_np * 255).convert("RGB")
    image = io.imread(image_name)
    imo = im.resize((image.shape[1], image.shape[0]))
    pb_np = np.array(imo)
    if type == "image":
        # Make and apply mask
        mask = pb_np[:, :, 0]
        mask = np.expand_dims(mask, axis=2)
        imo = np.concatenate((image, mask), axis=2)
        imo = Image.fromarray(imo, "RGBA")
    imo.save(d_dir + output_name)
# Remove Background From Image (Generate Mask, and Final Results)
def removeBg(imagePath):
    inputs_dir = os.path.join(currentDir, "static/inputs/")
    results_dir = os.path.join(currentDir, "static/results/")
    masks_dir = os.path.join(currentDir, "static/masks/")
    # convert string of image data to uint8
    with open(imagePath, "rb") as image:
        f = image.read()
        img = bytearray(f)
    nparr = np.frombuffer(img, np.uint8)
    if len(nparr) == 0:
        return "---Empty image---"
    # decode image
    try:
        img = cv2.imdecode(nparr, cv2.IMREAD_COLOR)
    except:
        # build a response dict to send back to client
        return "---Empty image---"
    # save image to inputs
    unique_filename = str(uuid.uuid4())
    cv2.imwrite(inputs_dir + unique_filename + ".jpg", img)
    # processing
    image = transform.resize(img, (320, 320), mode="constant")
    tmpImg = np.zeros((image.shape[0], image.shape[1], 3))
    tmpImg[:, :, 0] = (image[:, :, 0] - 0.485) / 0.229
    tmpImg[:, :, 1] = (image[:, :, 1] - 0.456) / 0.224
    tmpImg[:, :, 2] = (image[:, :, 2] - 0.406) / 0.225
    tmpImg = tmpImg.transpose((2, 0, 1))
    tmpImg = np.expand_dims(tmpImg, 0)
    image = torch.from_numpy(tmpImg)
    image = image.type(torch.FloatTensor)
    image = Variable(image)
    d1, d2, d3, d4, d5, d6, d7 = net(image)
    pred = d1[:, 0, :, :]
    ma = torch.max(pred)
    mi = torch.min(pred)
    dn = (pred - mi) / (ma - mi)
    pred = dn
    save_output(
        inputs_dir + unique_filename + ".jpg",
        unique_filename + ".png",
        pred,
        results_dir,
        "image",
    )
    save_output(
        inputs_dir + unique_filename + ".jpg",
        unique_filename + ".png",
        pred,
        masks_dir,
        "mask",
    )
    return "---Success---"
# ------- Load Trained Model --------
print("---Loading Model---")
model_name = "u2net"
model_dir = os.path.join(currentDir, "saved_models", model_name, model_name + ".pth")
net = U2NET(3, 1)
if torch.cuda.is_available():
    net.load_state_dict(torch.load(model_dir))
    net.cuda()
else:
    net.load_state_dict(torch.load(model_dir, map_location="cpu"))
# ------- Load Trained Model --------
print("---Removing Background...")
# ------- Call The removeBg Function --------
imgPath = "Image_File_Path"  # Change this to your image path
print(removeBg(imgPath))

Scroll down to the end of the script and add your desired image’s path to imgPath and run the code in the terminal by typing python __init__.py.

For example, I'll use the following image of a car.

Sample image with background

I'll add its path to imgPath test our code.

Adding image’s path

Now, run the code in the terminal by running python __init__.py.

Running the code in terminal

As a result, it will create the following image inside the project folder.

Resulting image with background removed

Awesome! Our code is working fine. For comparison, we can try the same image in other tools. The results will be pretty similar.

Coding the API

You can make a few simple changes to the code to turn it into a sellable API! Here’s what the code will look like after making the changes.

python
Copy
import torch
import torch.nn as nn
import torch.optim as optim
import numpy as np
import cv2
import uuid
import os
from model import U2NET
from torch.autograd import Variable
from skimage import io, transform
from PIL import Image
from fastapi import FastAPI, File, UploadFile
from fastapi.responses import JSONResponse
app = FastAPI()
# Get The Current Directory
currentDir = os.path.dirname(__file__)
# Functions:
# Save Results
def save_output(image_name, output_name, pred, d_dir, type):
    predict = pred
    predict = predict.squeeze()
    predict_np = predict.cpu().data.numpy()
    im = Image.fromarray(predict_np * 255).convert("RGB")
    image = io.imread(image_name)
    imo = im.resize((image.shape[1], image.shape[0]))
    pb_np = np.array(imo)
    if type == "image":
        # Make and apply mask
        mask = pb_np[:, :, 0]
        mask = np.expand_dims(mask, axis=2)
        imo = np.concatenate((image, mask), axis=2)
        imo = Image.fromarray(imo, "RGBA")
    imo.save(d_dir + output_name)
# Remove Background From Image (Generate Mask, and Final Results)
@app.get("/removeBG/")
async def removeBG(image_file: str):
    # ------- Load Trained Model --------
    print("---Loading Model---")
    model_name = "u2net"
    model_dir = os.path.join(
        currentDir, "saved_models", model_name, model_name + ".pth"
    )
    net = U2NET(3, 1)
    if torch.cuda.is_available():
        net.load_state_dict(torch.load(model_dir))
        net.cuda()
    else:
        net.load_state_dict(torch.load(model_dir, map_location="cpu"))
    # ------- Load Trained Model --------
    inputs_dir = os.path.join(currentDir, "static/inputs/")
    results_dir = os.path.join(currentDir, "static/results/")
    masks_dir = os.path.join(currentDir, "static/masks/")
    # convert string of image data to uint8
    with open(image_file, "rb") as image:
        f = image.read()
        img = bytearray(f)
    nparr = np.frombuffer(img, np.uint8)
    if len(nparr) == 0:
        return "---Empty image---"
    # decode image
    try:
        img = cv2.imdecode(nparr, cv2.IMREAD_COLOR)
    except:
        # build a response dict to send back to client
        return "---Empty image---"
    # save image to inputs
    unique_filename = str(uuid.uuid4())
    cv2.imwrite(inputs_dir + unique_filename + ".jpg", img)
    # processing
    image = transform.resize(img, (320, 320), mode="constant")
    tmpImg = np.zeros((image.shape[0], image.shape[1], 3))
    tmpImg[:, :, 0] = (image[:, :, 0] - 0.485) / 0.229
    tmpImg[:, :, 1] = (image[:, :, 1] - 0.456) / 0.224
    tmpImg[:, :, 2] = (image[:, :, 2] - 0.406) / 0.225
    tmpImg = tmpImg.transpose((2, 0, 1))
    tmpImg = np.expand_dims(tmpImg, 0)
    image = torch.from_numpy(tmpImg)
    image = image.type(torch.FloatTensor)
    image = Variable(image)
    d1, d2, d3, d4, d5, d6, d7 = net(image)
    pred = d1[:, 0, :, :]
    ma = torch.max(pred)
    mi = torch.min(pred)
    dn = (pred - mi) / (ma - mi)
    pred = dn
    save_output(
        inputs_dir + unique_filename + ".jpg",
        unique_filename + ".png",
        pred,
        results_dir,
        "image",
    )
    save_output(
        inputs_dir + unique_filename + ".jpg",
        unique_filename + ".png",
        pred,
        masks_dir,
        "mask",
    )
    return "---Success---"
# print("---Removing Background...")
# ------- Call The removeBg Function --------
# imgPath = "1.jpg"  # Change this to your image path
# print(removeBg(imgPath))

So, now you can deploy this on a server like Heroku, and voila! Your very own API is ready to be sold on API marketplaces.

Sell on Rapid API Hub

The best place to sell your API is the Rapid API Hub, the largest API directory in the world, where you can list your APIs and sell access to users who want to use them.

RapidAPI Hub

With different plans available, you can control how you monetize your creation and turn it into a lucrative business opportunity.

Wrap up

In summary, this course presents an incredible opportunity to create and sell an API that removes backgrounds from images. The process is simple, quick, and, most importantly, profitable! So, what are you waiting for? Get started today and let your creativity flow while making a handsome income!