OpenCV kyauta ce ta dakin karatun hangen nesa (nau'ikan da ke akwai na GNU / Linux, Mac OS X, Windows da Android) wanda ya kasance asali ci gaba ta Intel kuma ana amfani da shi a aikace-aikace marasa ƙimadaga tsarin tsaro tare da gano motsi, don aiwatar da aikace-aikacen sarrafawa inda ake buƙatar gane abu. Wannan saboda an bayar da bugu a ƙarƙashin lasisin BSD, wanda ke ba da damar amfani da shi kyauta don kasuwanci da dalilai na bincike tare da yanayin da aka bayyana a ciki.
Bude CV ya ƙunshi ayyuka sama da 500 waɗanda ke rufe ɗumbin wurare a cikin aikin hangen nesa, kamar gane abu (fitowar mutum), gyaran kamara, hangen nesa, hangen nesa na mutumtaka, rarraba ayyuka a cikin bidiyo, canza hotuna, cire samfuran 3D, ƙirƙirar sarari 3D daga hoto na kyamarar sitiriyo wanda ke ƙirƙirar hotuna masu inganci ta hanyar haɗa hotuna low quality.
Hakanan yana ba da damar bincika hotunan abubuwa iri ɗaya zuwa saitin abubuwan da aka gabatar ta hanyar amfani da hanyoyin koyon na'ura, tsara alamomi, gano abubuwan gama gari a hotuna daban-daban, kawar da lahani ta atomatik kamar jajayen idanu.
OpenCV yana samar da algorithms sama da 2500, duka na gargajiya da kuma tunowa da sabbin nasarori a fagen hangen nesa na kwamfuta da tsarin koyon na'ura. An rubuta lambar laburaren a cikin C ++ kuma an rarraba ta ƙarƙashin lasisin BSD.
Game da sabon sigar OpenCV 4.2
A halin yanzu laburaren yana cikin sigar OpenCV 4.2, a ciki a cikin tsarin DNN (Deep neural network) tare da aiwatar da tsarin koyar da aikin injiniya bisa tsarin hanyoyin sadarwa, kara da baya don amfani da CUDA kuma an aiwatar da tallafi na gwaji don nGraph OpenVINO API.
Baya ga yin amfani da umarnin SIMD, muna haɓaka aikin kod don fitowar sitiriyo (StereoBM / StereoSGBM), sake girman kansa, rufe fuska, juyawa, lissafin abubuwan launuka da suka ɓace, da sauran ayyukan da yawa.
A cikin tsarin G-API (budecv_gapi), wanda ke aiki a matsayin injin sarrafawa ingantaccen hoto ta amfani da zane-zane mai amfani da zane-zane, yana tallafawa haɓakar haɓakar algorithms mai haɗari don hangen nesa na kwamfuta da kuma zurfin koyon inji. Yana bayar da tallafi ga Ingin Inference Intel. Ara tallafi don sarrafa rafin bidiyo zuwa ƙirar aiwatarwa.
Canjin yanayin aiki (CVE-2019-5063, CVE-2019-5064) wanda zai iya haifar da aiwatar da lambar hari ta hanyar sarrafa bayanan da ba a tantance su ba a cikin tsarin XML, YAML da JSON. Idan ana samun hali tare da alamar mara amfani a yayin binciken JSON, ana kwafin dukkan ƙimar zuwa maɓallin, amma ba tare da tabbataccen ƙayyadadden iyakar yankin ƙwaƙwalwar da aka ware ba.
Na sauran canje-canje gabatar a cikin wannan sabon sigar:
- Addara aiwatar da ayyukan pyrDown da yawa.
- Ara ikon cire rafukan bidiyo daga kwantena na kafofin watsa labarai (demuxing) ta amfani da tushen bidiyo na tushen FFmpeg.
- Ara wani algorithm don sake zaɓin mitar saurin azabtarwar hotunan FSR (Sake Tsara Yanayin Mita).
- Methodara hanyar RIC don yin musayar wurare marasa kan gado.
- Methodara hanyar daidaita daidaitattun LOGOS.
Yadda ake girka OpenCV 4.2?
Ga waɗanda ke da sha'awar iya shigar da wannan laburaren, iya samun sabon sigar haka nan kuma tuntuɓi bayanan da suka danganci amfani har ma da samun koyarwa daga tashar aikinta.
A cikin wannan labarin Za mu samar da matakai don iya aiwatar da laburaren a kan Rasberi pi.
Don shigar da OpenCV akan Rasberi Pdole ne in sami tsarinku, wanda yake Raspbian.
Daga vZa mu bude tashar mota kuma a ciki za mu buga wadannan umarni don girka masu dogaro, kayan aikin haɓaka, fakitin hoto tsakanin sauran ɗakunan karatu:
sudo apt-get install build-essential cmake pkg-config
sudo apt-get install libjpeg-dev libtiff5-dev libjasper-dev libpng-dev libavcodec-dev libavformat-dev libswscale-dev libv4l-dev libxvidcore-dev libx264-dev libfontconfig1-dev libcairo2-dev libgdk-pixbuf2.0-dev libpango1.0-dev libgtk2.0-dev libgtk-3-dev libatlas-base-dev gfortran libhdf5-dev libhdf5-serial-dev libhdf5-103 libqtgui4 libqtwebkit4 libqt4-test python3-pyqt5
A ƙarshe, Bari mu shigar da fayilolin taken Python 3 don haka zamu iya tattara OpenCV:
sudo apt-get install python3-dev
Yanzu bari mu kirkiro yanayin Python tare da umarni masu zuwa, wannan don samun keɓaɓɓun rukunin yanar gizo:
wget https://bootstrap.pypa.io/get-pip.py
sudo python get-pip.py
sudo python3 get-pip.py
sudo rm -rf ~/.cache/pip
Za mu shigar da kyawawan dabi'u da kamala:
sudo pip install virtualenv virtualenvwrapper
nano ~/.bashrc
# virtualenv and virtualenvwrapper
export WORKON_HOME=$HOME/.virtualenvs
export VIRTUALENVWRAPPER_PYTHON=/usr/bin/python3
source /usr/local/bin/virtualenvwrapper.sh
source ~/.bashrc
mkvirtualenv cv -p python3
pip install "picamera[array]"
Anyi wannan yanzu zamu tattara budeV tare da:
cd ~
wget -O opencv.zip https://github.com/opencv/opencv/archive/4.2.0.zip
wget -O opencv_contrib.zip https://github.com/opencv/opencv_contrib/archive/4.2.0.zip
unzip opencv.zip
unzip opencv_contrib.zip
mv opencv-4.2.0 opencv
mv opencv_contrib-4.2.0 opencv_contrib
Yanzu zamu kara canzawa a cikin tsarin mu tunda idan muka barshi kamar yadda yake ta yadda ake so tsarin zai iya rataye:
sudo nano /etc/dphys-swapfile
Kuma zamu shirya mai canzawa CONF_SWAPSIZE:
CONF_SWAPSIZE=1024
Muna adanawa da rufewa tare da ctrl + o da ctrl + x. Sannan mu rubuta:
sudo /etc/init.d/dphys-swapfile stop
sudo /etc/init.d/dphys-swapfile start
Yanzu zamu ci gaba da tattarawa:
workon cv
pip install numpy
cd ~/opencv
mkdir build
cd build
cmake -D CMAKE_BUILD_TYPE=RELEASE \
-D CMAKE_INSTALL_PREFIX=/usr/local \
-D OPENCV_EXTRA_MODULES_PATH=~/opencv_contrib/modules \
-D ENABLE_NEON=ON \
-D ENABLE_VFPV3=ON \
-D BUILD_TESTS=OFF \
-D INSTALL_PYTHON_EXAMPLES=OFF \
-D OPENCV_ENABLE_NONFREE=ON \
-D CMAKE_SHARED_LINKER_FLAGS=-latomic \
-D BUILD_EXAMPLES=OFF ..
make -j4
sudo make install
sudo ldconfig
cd /usr/local/lib/python3.7/site-packages/cv2/python-3.7
sudo mv cv2.cpython-37m-arm-linux-gnueabihf.so cv2.so
cd ~/.virtualenvs/cv/lib/python3.7/site-packages/
ln -s /usr/local/lib/python3.7/site-packages/cv2/python-3.7/cv2.so cv2.so
Kuma a shirye.