bug: bagOfFeatures Internal calculation of array dimension match error

ENV: matlab 2022b, win10

When I "bagOfFeatures" some of my own images, the following error occurs.

parkingLotRoot = "./parkingLotImages";
imds = imageDatastore(parkingLotRoot);

bofData = bagOfFeatures(imds,CustomExtractor=@helperORBFeatureExtractorFunction,...
    TreeProperties=[5,10]);

Unable to perform assignment because the size of the left side is 1-by-1 and the size of the right side is 32-by-1. Error in vision.internal.kmeans.DistanceMetricHamming/dist (line 30) d(:,1) = sum(this.LUT(bitvec+1), 2); Error in vision.internal.kmeans.ClusterCompactness/clusterCompactness (line 23) d = dist(this.Metric, features, centers(assignments,:)); Error in vision.internal.kmeans.clusterdata (line 134) prevCompactness = clusterCompactness(compactnessMetric, features, centers, assignments); Error in vision.internal.bof.VocabularyBuilderKMedians/create (line 42) clusterCenters = vision.internal.kmeans.clusterdata(... Error in vision.internal.bof.VocabularyBuilderHierarchicalClustering/invokeVocabularyBuilder (line 121) words = create(this.VocabularyBuilder, features, branchingFactor); Error in vision.internal.bof.VocabularyBuilderHierarchicalClustering/buildVocabularyTree (line 105) vt.Words = invokeVocabularyBuilder(this, features, printer); Error in vision.internal.bof.VocabularyBuilderHierarchicalClustering/buildVocabularyTree (line 98) [vt.Nodes(i), step] = buildVocabularyTree(this, fgroup, level-1, i, step, printer); Error in vision.internal.bof.VocabularyBuilderHierarchicalClustering/buildVocabularyTree (line 98) [vt.Nodes(i), step] = buildVocabularyTree(this, fgroup, level-1, i, step, printer); Error in vision.internal.bof.VocabularyBuilderHierarchicalClustering/buildVocabularyTree (line 98) [vt.Nodes(i), step] = buildVocabularyTree(this, fgroup, level-1, i, step, printer); Error in vision.internal.bof.VocabularyBuilderHierarchicalClustering/buildVocabularyTree (line 98) [vt.Nodes(i), step] = buildVocabularyTree(this, fgroup, level-1, i, step, printer); Error in vision.internal.bof.VocabularyBuilderHierarchicalClustering/create (line 56) vt = buildVocabularyTree(this, features, this.NumLevels,[], step, printer); Error in bagOfFeatures/createVocabulary (line 793) vocabulary = this.VocabularyBuilder.create(descriptors); Error in bagOfFeatures (line 301) vocabulary = createVocabulary(this, descriptorSet); Error in testDebug (line 11) bofData = bagOfFeatures(imds,CustomExtractor=@helperORBFeatureExtractorFunction,...

how can i solve it ,your answer would be greatly appreciate! To reproduce the bug, since matlab answer attachments cannot be uploaded to files larger than 5Mb, I am posting them here.

https://ww2.mathworks.cn/matlabcentral/fileexchange/121198-bagoffeatures-bug-report-images

support function is here:

function [features, featureMetrics]= helperORBFeatureExtractorFunction(I)
% helperORBFeatureExtractorFunction Implements the ORB feature extraction 
% used in bagOfFeatures.
%
%   This is an example helper function that is subject to change or removal 
%   in future releases.

%   Copyright 2021 The MathWorks, Inc.

numPoints   = 1000;

% Detect ORB features
Igray  = im2gray(I);

points = detectORBFeatures(Igray, 'ScaleFactor', 1.2, 'NumLevels', 8);

% Select a subset of features, uniformly distributed throughout the image
points = selectUniform(points, numPoints, size(Igray, 1:2));

% Extract features
features = extractFeatures(Igray, points);

% Compute the Feature Metric. Use the variance of features as the metric
featureMetrics = var(single(features.Features),[],2);
end

[Qu Cao]

Thank you for reporting this! We will fix the bug in a future update. Meanwhile, do you mind trying the following workaround?

bofData = bagOfFeatures(imds,CustomExtractor=@helperORBFeatureExtractorFunction,...

TreeProperties=[4,10]);

It looks to me that a smaller value will make it work.

[xingxingcui]

You are right, changing to a smaller value is not buggy,多谢曹博~