Download My custom trained Opencv Cascade Classifier for detectMultiScale

This tutorial contains a list of custom trained LBP and HAAR cascade trained for Opencv CascadeClassifier detect multiscale method. You can download and test these cascades to detect people, heads, and cars in OpenCV. They are of mixed quality and mainly tuned for performance. I preferred LBP exactly for the performance reason. The cascades are not perfect but available for free. The cascades are useful in many projects.  The training data are originally mine as well. Please cite the blog if you are using these cascades. Thanks

I created the car dataset from the video from my prone. The base set of positive car samples was very poor. Basically, I used background subtraction to extract moving bounding boxes over the highway. Then the dataset was clean manually. The duplicates and very similar pictures were removed for better robustness of the detector. 

Opencv car cascade classifier code

Opencv Head Cascade Classifier

Head cascade download HERE
To create this dataset was very tricky. The problem was to isolate the heads. I have a dataset for the upper body as a result of the crop the upper 30 percent of pedestrian and people detection. I want to train the detector more focused just to head detection. It can achieve some benefit in multi-target tracking. I perform the collection of the head purely manually. I dataset contains more than 2000 heads as positive samples and some random pictures as negatives.

Opencv head Cascade Classifier properties

Opencv People Cascade Classifier

People cascade download link
People cascade 2 download
This cascades was trained on the people positive samples extracted mostly from security cameras view. The process of dataset collection was the same as well in car case. The background substraction collect very messy collection of moving people in foreground. This was clear manually to extract dataset of various situation, where the cuplicates are removed for better robustness.

Opencv people Cascade Classifier properties 1

The second people cascade Cascade Classifier properties

<?xml version="1.0"?>

<!--

This is just basic 12 stage haar cascade pedestrian detector develop by 

V.K. from https://funvision.blogspot.com

-->

<opencv_storage>

<cascade>

  <stageType>BOOST</stageType>

  <featureType>LBP</featureType>

  <height>64</height>

  <width>32</width>

  <stageParams>

    <boostType>GAB</boostType>

    <minHitRate>9.9500000476837158e-01</minHitRate>

    <maxFalseAlarm>4.1999998688697815e-01</maxFalseAlarm>

    <weightTrimRate>9.4999999999999996e-01</weightTrimRate>

    <maxDepth>1</maxDepth>

    <maxWeakCount>100</maxWeakCount></stageParams>

  <featureParams>

    <maxCatCount>256</maxCatCount>

    <featSize>1</featSize></featureParams>

  <stageNum>13</stageNum>

Opencv Cascade classifier performance and detection considerations

Code how to use opencv Cascade Classifier for detectMultiScale

//V.K. from https://funvision.blogspot.com

// Name of the downloaded my cascades.. 

 string cascadeHead = "cascadeH5.xml";

 string cascadeName = "cascadG.xml";



// Load the cascade

 CascadeClassifier detectorBody;

 bool loaded1 = detectorBody.load(cascadeName);

 CascadeClassifier detectorHead;

 bool loaded2 = detectorHead.load(cascadeHead);

        

// save original make img gray

// draw rectangle back to the original colored sample

 Mat original;

 img.copyTo(original);

// Prepare vector for results 

vector<Rect> human;

vector<Rect> head;

// Prepare gray image

 cvtColor(img, img, CV_BGR2GRAY);

// equalize Histogram  

        equalizeHist(img, img);

// detect body and head in the img 

// Set the proper min and max size for your case

 detectorBody.detectMultiScale(img, human, 1.04, 4, 0 | 1, Size(30, 80), Size(80,200));

 detectorHead.detectMultiScale(img, head, 1.1, 4, 0 | 1, Size(40, 40), Size(100, 100));

 if (human.size() > 0) {

  for (int gg = 0; gg < human.size(); gg++) {

   rectangle(original, human[gg].tl(), human[gg].br(), Scalar(0, 0, 255), 2, 8, 0);

     }

  }

 if (head.size() > 0) {

  for (int gg = 0; gg < head.size(); gg++) {

   rectangle(original, head[gg].tl(), head[gg].br(), Scalar(0, 0, 255), 2, 8, 0);

     }

 }

Opencv LBP HAAR Cascade Classifier for detectMultiScale in action