Fred'ô farias06

C’est pas gagné pour faire la détection du contour du blob (en mv4) :

[video width=“1060” height=“720” m4v=“https://www.cyber-neurones.org/wp-content/uploads/2022/04/blob2outpytest.m4v"][/video]

J’ai des problèmes de jaune :(

import cv2
import numpy as np 
import matplotlib.pyplot as plt
....
        frame_hsv=cv2.cvtColor(frame, cv2.COLOR_BGR2HSV)

        low_yellow=np.array([10, 125, 100])
        upper_yellow=np.array([60, 220, 220])
        mask_yellow=cv2.inRange(frame_hsv,low_yellow,upper_yellow)

        list_masks=[mask_yellow]
        for i,mask in enumerate(list_masks):

            contours, _=cv2.findContours(mask,cv2.RETR_LIST,cv2.CHAIN_APPROX_NONE)

Sans le polygone (et en mp4):

[video width=“1280” height=“720” mp4=“https://www.cyber-neurones.org/wp-content/uploads/2022/04/blob2outpytest2.mp4"][/video]

A suivre.

J’essaye donc un cinquième projet : https://github.com/adipandas/multi-object-tracker . Un peu complexe a installer mais vu que j’avais déjà fait le gros du travail dans mes précédents post :

• https://www.cyber-neurones.org/2020/10/draft-ubuntu-jupyter-notebook-python3-tensorflow-opencv-comptage-des-cyclistes/ : (Draft) Ubuntu, Jupyter Notebook, Python3, Tensorflow, OpenCV : Comptage des cyclistes
• https://www.cyber-neurones.org/2020/10/draft-ubuntu-jupyter-notebook-python3-tensorflow-opencv-detection-des-cyclistes-et-des-casques/ : (Draft) Ubuntu, Jupyter Notebook, Python3, Tensorflow, OpenCV : Détection des cyclistes et des casques

Ce projet est bien interessant car il fonctionne avec ssd_mobilenet_v2_coco_2018_03_29 . Alors que dans le précédent j’étais avec la version ssd_mobilenet_v1_coco .

Je fais un petit brouillon de mon installation :

• Ubuntu 18.04.4 LTS
• Kernel 5.4.0-42-generic (uname -a)
• Python 3.6.9 (python3 –version)
• Jupyter Notebook 6.1.4
• OpenCV : 3.2.0

ii  libopencv-calib3d3.2:amd64         3.2.0+dfsg-4ubuntu0.1   amd64        computer vision Camera Calibration library
ii  libopencv-contrib3.2:amd64         3.2.0+dfsg-4ubuntu0.1   amd64        computer vision contrlib library
ii  libopencv-core3.2:amd64            3.2.0+dfsg-4ubuntu0.1   amd64        computer vision core library
ii  libopencv-features2d3.2:amd64      3.2.0+dfsg-4ubuntu0.1   amd64        computer vision Feature Detection and Descriptor Extraction library
ii  libopencv-flann3.2:amd64           3.2.0+dfsg-4ubuntu0.1   amd64        computer vision Clustering and Search in Multi-Dimensional spaces library
ii  libopencv-highgui3.2:amd64                    3.2.0+dfsg-4ubuntu0.1   amd64        computer vision High-level GUI and Media I/O library
ii  libopencv-imgcodecs3.2:amd64                  3.2.0+dfsg-4ubuntu0.1   amd64        computer vision Image Codecs library
ii  libopencv-imgproc3.2:amd64                    3.2.0+dfsg-4ubuntu0.1   amd64        computer vision Image Processing library
ii  libopencv-ml3.2:amd64                         3.2.0+dfsg-4ubuntu0.1   amd64        computer vision Machine Learning library
ii  libopencv-objdetect3.2:amd64                  3.2.0+dfsg-4ubuntu0.1   amd64        computer vision Object Detection library
ii  libopencv-photo3.2:amd64                      3.2.0+dfsg-4ubuntu0.1   amd64        computer vision computational photography library
ii  libopencv-shape3.2:amd64                      3.2.0+dfsg-4ubuntu0.1   amd64        computer vision shape descriptors and matchers library
ii  libopencv-stitching3.2:amd64      3.2.0+dfsg-4ubuntu0.1   amd64        computer vision image stitching library
ii  libopencv-superres3.2:amd64       3.2.0+dfsg-4ubuntu0.1   amd64        computer vision Super Resolution library
ii  libopencv-video3.2:amd64          3.2.0+dfsg-4ubuntu0.1   amd64        computer vision Video analysis library
ii  libopencv-videoio3.2:amd64        3.2.0+dfsg-4ubuntu0.1   amd64        computer vision Video I/O library
ii  libopencv-videostab3.2:amd64      3.2.0+dfsg-4ubuntu0.1   amd64        computer vision video stabilization library
ii  libopencv-viz3.2:amd64            3.2.0+dfsg-4ubuntu0.1   amd64        computer vision 3D data visualization library

• TensorFlow : 1.14.0 ( python3 -c ‘import tensorflow as tf; print(tf.__version__)’  )
• TensorFlow : 2.3.1 :

pip3 show tensorflow
Name: tensorflow
Version: 2.3.1
Summary: TensorFlow is an open source machine learning framework for everyone.
Home-page: https://www.tensorflow.org/
Author: Google Inc.
Author-email: packages@tensorflow.org
License: Apache 2.0
Location: /home/arias/.local/lib/python3.8/site-packages
Requires: tensorboard, gast, protobuf, termcolor, absl-py, numpy, tensorflow-estimator, opt-einsum, wheel, astunparse, h5py, google-pasta, wrapt, grpcio, six, keras-preprocessing
Required-by: 

Les projet github utilisés :

git clone https://github.com/iamprahladsinghnegi/automaticDetectionOfBikeRiderDrivingWithoutHelmet.git 
cd automaticDetectionOfBikeRiderDrivingWithoutHelmet
git clone https://github.com/tensorflow/models 
cp -r inference_graph/ models/research/object_detection/.
cp -r test1.jpg output.jpg models/research/object_detection/.
cp -r training/ models/research/object_detection/
cp automaticDetectionOfBikeRiderDrivingWithoutHelmet.ipynb models/research/object_detection/.

Installation (dans le désordre):

sudo apt-get -y install python3python-pip python-dev
sudo -H pip3 install jupyter
pip3 install numpy
sudo apt-get install python3-google-apputils
sudo apt install python3-tensorflow
pip3 install tensorflow-cpu
python3 -m pip install tensorflow
python3 -m pip install utils
sudo apt  install protobuf-compiler
python3 -m pip install matplotlib

Important :

Une fois que l’on a les deux projets il faut faire la commande ( dans automaticDetectionOfBikeRiderDrivingWithoutHelmet/models/research/ ):