preprocess - Preprocess training and test images - MATLAB (original) (raw)

Preprocess training and test images

Since R2021a

Syntax

Description

[outputData](#mw%5Fdf2fa24d-e8a3-4f07-90af-cf0baf7883cd) = preprocess([detector](#mw%5F2f18f1b4-b974-49ce-bdfb-97387b91859b),[trainingData](#mw%5F85dce6f5-b7bc-4429-bf28-6ee56342ed01)) preprocesses the training data trainingData before using it to train the YOLO v3 object detector. The training images and the corresponding bounding boxes are stored in the trainingData. The preprocess function performs these operations:

Rescales the intensity values of the training images to the range [0, 1].
Resizes the training images to one of the nearest network input sizes and updates the bounding box coordinate values for accurate training. The function preserves the original aspect ratio of the training data.

example

[outputImg](#mw%5F1b074622-c59d-43a1-a997-610a6b824e74) = preprocess([detector](#mw%5F2f18f1b4-b974-49ce-bdfb-97387b91859b),[img](#mw%5F1ff43d0c-3630-40ab-a15f-cb46bccc6f51)) preprocesses the test images img for object detection using a YOLO v3 object detector. The preprocess function performs these operations:

Rescales the intensity values of the test images to the range [0, 1].
Resizes the test images to one of the nearest network input sizes and preserves the original aspect ratio of each test image.

example

[___,[scaleInfo](#mw%5Ff47453fd-9609-4044-8536-079e848a779f)] = preprocess(___) returns information on the scale factor applied for image resizing, in addition to any combination of arguments from previous syntaxes.

Examples

collapse all

Load a pretrained YOLO v3 object detector.

detector = yolov3ObjectDetector('tiny-yolov3-coco');

Load the training dataset into the workspace. The training data is a cell array that contains three training images and the corresponding bounding box values and class labels.

load('trainingData.mat','trainingData');

Resize the training images to the nearest network input size and rescale the intensity values by using the preprocess function.

[outputData,info] = preprocess(detector,trainingData);

Display the output images and the scale information used for resizing the images.

figure montage({outputData{1,1},outputData{2,1},outputData{3,1}},Size=[1 3]) title("Preprocessed Output Image")

Figure contains an axes object. The hidden axes object with title Preprocessed Output Image contains an object of type image.

ScaleX = [info{1,1}.ScaleX;info{2,1}.ScaleX;info{3,1}.ScaleX]; ScaleY = [info{1,1}.ScaleY;info{2,1}.ScaleY;info{3,1}.ScaleY]; table(ScaleX,ScaleY)

ans=3×2 table ScaleX ScaleY
_________ _________

0.0072115    0.0072115
0.0072115    0.0072115
0.0072115    0.0072115

Display the input and the preprocessed image size and bounding box values.

bboxIn = cell2table(trainingData,'VariableNames',{'Images','Bounding Boxes','Labels'})

bboxIn=3×3 table Images Bounding Boxes Labels
_________________ ________________________ ___________

{224×399×3 uint8}    220    136     35     28    {'vehicle'}
{224×399×3 uint8}    175    126     61     45    {'vehicle'}
{224×399×3 uint8}    108    120     45     33    {'vehicle'}

bboxOut = cell2table(outputData,'VariableNames',{'Images','Bounding Boxes','Labels'})

bboxOut=3×3 table Images Bounding Boxes Labels
__________________ ________________________ ___________

{416×416×3 single}    229    232     36     29    {'vehicle'}
{416×416×3 single}    182    222     64     46    {'vehicle'}
{416×416×3 single}    112    215     47     35    {'vehicle'}

Load a pretrained YOLO v3 object detector.

detector = yolov3ObjectDetector('tiny-yolov3-coco');

Read a test image.

I = imread('highway.png');

Resize the test image to the network input size and rescale the intensity values by using the preprocess function.

[outputImg,scaleInfo] = preprocess(detector,I);

Display the output image and the scale information used for resizing the image.

Figure contains an axes object. The hidden axes object contains an object of type image.

               ScaleX: 0.7692
               ScaleY: 0.5769
PreprocessedImageSize: [416 416 3]

Input Arguments

collapse all

Training data for YOLO v3 object detector, specified as a _N_-by-3 cell array that contains the images, bounding boxes, and the class labels. Each row is of the form [images bounding boxes _labels_]. N is the number of output layers in the network. The bounding boxes must be stored as a _K_-by-4 matrix of the form [x y width _height_]. K is the number of object classes.

Test images, specified as a numeric array of size_M_-by-_N_-by-C or_M_-by-_N_-by-C_-by-T.M is the number of rows, N is the number of columns, and C is the number of color channels. The value of_C is 1 for grayscale images and3 for RGB color images. T is the number of test images in the array.

Output Arguments

collapse all

Preprocessed training data, returned as a _N_-by-3 cell array.

Data Types: cell

Preprocessed test images, returned as a numeric array of size_P_-by-_Q_-by-C or_P_-by-_Q_-by-_C_-by-T.P and Q are the number of rows and columns, respectively, in the preprocessed image.

Information about the scale factor for resizing the input images, returned as a structure with fields PreprocessedImageSize,ScaleX, and ScaleY.

PreprocessedImageSize — Size of the output resized image.
ScaleX — Scale factor for resizing an image in the_X_-direction (along the rows).
ScaleY — Scale factor for resizing an image in the_Y_-direction (along the columns).

Data Types: struct

Version History

Introduced in R2021a