OCR not detecting numbers

18 次查看（过去 30 天）
Chris
Chris 2025-3-26
回答： Image Analyst 2025-3-28
I'm trying to use OCR to read numbers from different photos. I have pre-processed the images as seen below, and it just won't detect some even though they look very reading. I've put the pre-processed photos below to show which ones are detected and which aren't.
Able to detect:
Not able to detect:
Also not detected:

请先登录，再回答此问题。

回答（1 个）

Image Analyst
Image Analyst 2025-3-28
Ran in:
Maybe it's too big. Try shrinking it down to smaller images:
%--------------------------------------------------
% Read in original RGB image and convert to grayscale.
rgbImage = imread('72.png');
% Convert to gray level
binaryImage = rgbImage(:, :, 1) < 128;
binaryImage = imresize(binaryImage, 0.1);
imshow(binaryImage);
impixelinfo;
axis on
%--------------------------------------------------
% Find the number
txt = ocr(binaryImage, 'CharacterSet', '0123456789', 'LayoutAnalysis','line')
txt =
ocrText with properties: Text: '72...' CharacterBoundingBoxes: [4x4 double] CharacterConfidences: [4x1 single] Words: {'72'} WordBoundingBoxes: [24 11 88 69] WordConfidences: 0.5639 TextLines: {'72'} TextLineBoundingBoxes: [24 11 88 69] TextLineConfidences: 0.5639
fprintf('Detected "%s".\n', txt.Words{1})
Detected "72".

类别

在 Help Center 和 File Exchange 中查找有关 Convert Image Type 的更多信息

标签

产品

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by