Python Meets Pixels: Excel Now Analyzes Images Like a Pro

Office-Watch.comLast updated: 16 August 2025

Microsoft 365,Microsoft Excel,Microsoft Office,Microsoft Office for Mac,Office 365

Excel, long known for its love affair with numbers, has now developed a taste for images—thanks to Python. Microsoft has rolled out a feature that lets users insert images into individual Excel cells and analyze them using Python code. Yes, the spreadsheet just got a little more artsy and a lot more tech-savvy.

Forget floating images hovering over cells like ghosts of clipart past. This update allows users to plop an image squarely into a single cell (no cell-merging acrobatics allowed). Once inside, the image can be called upon with the Python in Excel integration and subjected to various analytical or creative transformations.

How It Works (Spoiler: It’s Simpler Than Expected)

For this you’ll need Excel 365 with a business plan that supports the Python extension of Excel.

Insert an image using the new =IMAGE() function. Just keep it inside one cell, Excel likes its boundaries respected.
Open a Python cell using =PY( and reference the image’s cell with xl("A1") (or wherever the image lives).
Use Python (and its faithful sidekick, Pillow) to do something clever. Microsoft’s demo? A sharpness detector using grayscale, Laplacian filters, and variance. High variance = “Sharp.” Low variance = “Blurry.”
Hit Ctrl + Enter and—presto—the result appears like magic (or like solid coding).

The Python code

This is the sample code from Microsoft. It converts the image to grayscale, adds a Laplacian filter the classifies the image variance.

from PIL import Image 

import numpy as np 

from scipy.signal import convolve2d 

# Convert image to grayscale and array 

image = xl("A1") 

arr = np.array(image.convert("L"), dtype=np.float32) 

# Apply Laplacian filter 

laplacian = convolve2d(arr, [[0, 1, 0], [1, -4, 1], [0, 1, 0]], mode='same', boundary='symm') 

# Classify based on variance 

"Blurry" if np.var(laplacian) < 100 else "Sharp"

What Can One Do with This Sorcery?

Besides checking if a selfie is blurry or not, users can:

Adjust brightness and contrast.
Add filters (hello, ExcelGram).
Watermark documents without third-party tools.
Extract metadata like image dimensions, camera info, or creation date.
Create fun Python scripts to map pixel color distributions or even detect objects.

A Few Nerdy Details

Pillow library (PIL) does the heavy lifting—opening, analyzing, and filtering image data right inside Excel.
Users can control image resolution for performance purposes via File | Options | Advanced | Python in Excel. Options range from “Actual size” to “Small” (for the minimalist analyst).