yuanyantai Goto Github PK

aiot

Cutting-Edge Artificial Intelligence on the Jetson Nano used to decode Brain Waves (EEG) on the Edge

appledetection

A dataset for detecting healthy and damaged apples

attention-gated-networks

Use of Attention Gates in a Convolutional Neural Network / Medical Image Classification and Segmentation

backupjetson

Backup a NVIDIA Jetson Developer Kit root system

bacnet-stack-stm32f4

Porting bacnet-stack to stm32f4

buildjetsonxaviernxkernel

Build the linux kernel and modules on board the Jetson Xavier NX.

buildopencvxavier

Build and install OpenCV for the NVIDIA Jetson AGX Xavier

canopenstm32

CANopenNode on STM32 microcontrollers.

chip-wafer-classification-deep-learning

This project uses AWS machine learning and IoT tools to develop a deep learning defect classification model and use it for real-time defect detection on a device.

coffee-cobra

Open source optical coffee sorter based on a Raspberry Pi 3.

coffee-quality-database

Building the Coffee Quality Institute Database

coffeebeans-classification

Coffee beans classification based on computational intelligence techniques

csharp-app-sdk

锐米 loraserver 应用 MQTT 通信数据SDK

curve-gcn

Official PyTorch code for Curve-GCN (CVPR 2019)

deep-learning-approach-for-surface-defect-detection

（最先进的缺陷检测网络） A Tensorflow implementation of "Segmentation-Based Deep-Learning Approach for Surface-Defect Detection"

deep_coffee

Machine Learning & Computer Vision for coffee beans selection

deep_object_pose

Deep Object Pose Estimation (DOPE) – ROS inference (CoRL 2018)

deeplearningexamples

Deep Learning Examples

deepstream_triton_model_deploy

How to deploy open source models using DeepStream and Triton Inference Server

defect-classifier

Final Year Project

defect-detection-classifier

Visual Defect Detection on Boiler Water Wall Tube Using Small Dataset

defect_classification

This code is for classifying healthy vs defective material.

defect_classifier

docker-tegra-ubuntu

Ubuntu for Tegra with Nvidia apt sources enabled

embedded_videos

Contents for my videos on embedded programming on Linux with STM32

embvision

Embedded Vision Tutorial with MATLAB(R)/Simulink(R) & Raspberry Pi(TM)

face

Face detection via webcam In daily life, seeing and hearing are used as the main sources of receiving Information from the environment. So, With the advancement of technology, new tools are needed for human communication, which similarly use visual or audio inputs. The advancement of new technologies in human-computer interaction is very impressive. Many researches has been done in this field proposed to the improvement of human-computer relationship. Recent technologies tend to interact with humans using body motions. To create an intuitive and natural interaction between humans and computers, we will use a camera for detecting faces. This section describes how to create and run a face detection algorithm in Python using OpenCV. In this part, we will fully explain the basic face recognition methodology, which includes cascading classification tools and deep learning CNNs. Cascade classification: OpenCv uses machine learning algorithms to find faces in the image. Since the faces are very complex, there is no test or pattern of simple properties that shows us whether a face has been found. Instead, there are thousands of small patterns and features that need to be matched with data. There are some algorithms that divide the face recognition process into thousands of smaller and smaller tasks that are so much easier to solve. These tasks and the process of classifying are called classifiers. You need 6000 for something like your face! Considering that all of them should be found in identifying a face (but with limited errors). But there is a problem: To detect the face the algorithm, starts from the left and above the image and moves downwards around the small blocks of data, looks at each block, and asks itself every time "It's a face ...? It's a face. .... This is a face? Since in each block, as discussed before, 6000 tests are performed on the block, you might have to make millions of calculations. To solve this problem, opencv uses cascades. The opencv cascade splits the facial recognition problem into several stages. For each block, a very accurate and fast test is done. OpenCv is the most popular and most used library for computer vision. Written in C ++ / C and has a version for Python. OpenCv uses machine learning algorithms to find components in the image. Installing OpenCv The first thing to do is to install the latest version of Python itself with Pip! To insure that opencv is installed or not, you can type in python in cmd to display it in CLI. Then add the module using import Opencv if you do not have the warning error you can make sure that it is installed. Running the code: For running the code its necessary to have an opencv in python. After running the code the camera of the laptop will be turned on and we can see a green rectangle around the faces that are in front of the camera.

halcon_licenses

Halcon licenses activation

hosts

:statue_of_liberty:最新可用的google hosts文件。镜像：

image-segmentation-keras

Implementation of Segnet, FCN, UNet , PSPNet and other models in Keras.