I. Preface

With the progress of the times, science and technology are changing with each passing day. Modern information technologies such as cloud computing, artificial intelligence, block chains, big data and so on have changed people's thinking, production, life and learning mode.

The development of video surveillance cloud computing technology promotes the development of large data in the Internet of Things to break through the "space-time and geographical limitations" and achieve instant vision, and make use of in-depth learning, video image recognition, large data depth processing (DT), sensors and other technologies to achieve video surveillance in advance perception, to move towards the interconnection of all things.

II. On Miling Cloud

Menning is a leading inter-regional and video-centric Internet of Things project developed by Shenzhen Shenchuangyuan Digital Technology Co., Ltd. It adopts the "cloud-tube-end-control" strategic design architecture model, cloud-side deep learning-cloud, low bit-stream compression transmission-tube, front-end convergence terminal-end, video object-connected large data wall processing-control. The aim is to provide decision support for the integrated management of large-scale, city-level, cross-regional and ITU sensors centralized video surveillance and intelligent distribution control.

Shenchuanyuan's Miling Cloud Platform has integrated hardware and software solutions from front-end to back-end, and provides open cloud platform services to B-end users and developers to quickly realize cross-regional video surveillance and Internet of Things connection.

Milling makes video surveillance easier and more efficient across regions. Milling Branches can assist users to deploy the overall solution of private cloud monitoring and public cloud monitoring, and let the video-based intelligent Internet of Things penetrate into all walks of life. We hope to bring convenience to society with products and services, let machines perceive the world, and embrace a new generation of intelligent change with partners.

3. Application Block Diagram of Miling Cloud Interconnection Monitoring Solution

IV. SOFTWARE AND HARDWARE PRODUCT PARAMETERS OF SOLUTIONS

(1) Cloud-Miling Cloud Platform

OpenStack cloud computing platform is used to help service providers and businesses to quickly implement video link cloud services (infrastructure as a service, laaS). It adopts high-performance deep learning architecture, has high-performance pull-through, penetration and so on, and can intelligently handle pull-over 1000-2000 network video data access at the same time. It has massive Internet of Things data access cloud platform, Web browser real-time monitoring traffic and PC client, mobile phone observation monitoring video data and Internet of Things data, timely display whether the upload node has device anomalies and logs, decision-making scheme to support users to deal with related early warning affairs, flat operation management software, real-time monitoring cloud data.

(2) Tube-Low Bit Stream Compression Transmission

"Meanning" is a leading domestic solution for remote video surveillance and Internet of Things. It breaks through the problems and technical barriers encountered in cross-regional video surveillance. It adopts the transmission of H.265 secondary low bit stream compression on the public network and penetrates the upper wall to the center of large data platform. It has the characteristics of better flow and real-time, and reduces the bandwidth requirement to the greatest extent. Intelligent distribution of video and Internet of Things data.

(3) "End" - Cloud Convergence Terminal

Compatible with major brands of network cameras, using independent Deep Learning secondary compression algorithm, low bit stream H.265 is uploaded to the cloud. It has excellent characteristics such as low latency, less bandwidth, automatic repair of broken lines, automatic remote upgrade, etc. It is also connected to Internet of Things sensors (PM2.5, fire protection, noise, water quality, illumination, pressure, PH value, switch volume, infrared and so on) through RS485 interface. Sensor) data can be transferred to the platform through MQTT, NB-IOT and other protocols to support Huawei LiteOS access platform.

Type:

(4) Control-Milling CDT Data Integrated Processing Platform

◆Introduction

Milling CDT data integrated processing platform uses the system architecture of FPGA + ARM to run embedded Linux system, which guarantees that the system can run stably in 7*24 hours environment. Modular design provides high flexibility and maintainability. FPGA completes video image processing and data transmission; ARM builds communication, control interface, video coding and decoding system and intelligent analysis modules, alarm module, greatly enhance the flexibility of the system, can achieve cross-platform interaction, support Anroid/IOS/Windows/Linux. It can display any window opening, roaming, overlapping, zooming, rotation and other display effects of the display screen. It also supports a variety of video source input modes, and also supports the access of local IPC video images. The platform uses bus switching technology, supports IPC decoding card, multimedia function card, intelligent analysis card, video structured, alarm card and other business cards, which well solves the user's demand for data processing. Pure embedded design and plug-in application greatly increase the stability, scalability and maintainability of the system.

Milling CDT Data Integrated Processing Platform

◆Cloud Video Decoding and Multimedia Technology

It supports stream decoding from external cloud and Internet of Things data wall, IP stream real-time decoding supports H.264/H.265, and supports local ONVIF, RTSP and other network streaming media protocols, compatible with mainstream brand network cameras (IPC), NVR, video streaming media servers, etc. A single network port supports 72-way D1/32-way 720/18-way 1080P/4-way 4K decoding.

Supports screen projection functions of mobile phones, tablets and laptops.

Multimedia Applications

◆ Video Image Processing

A single signal can be spliced and displayed on any M*N(>0) display unit.

Support input signal arbitrary splicing, window opening, roaming, support layer overlay, single screen support four layers.

Single screen supports 36 screen segmentation, which needs to be used in conjunction with IP card.

Intelligent scaling technology, image scaling adopts intelligent multi-phase filtering algorithm, according to the characteristics of the image automatically select the optimal filter coefficient, the image is close to losing magnitude.

Infinite zoom effect ensures more details, no jagged edges and good sharpness.

Super High Resolution Base Map Supporting 3840*2160 Pixels

Supporting the scrolling caption function of the LED, the LED caption needs to occupy a layer.

Multi-platform client unified management, automatic detection of input and output signals and business cards.

It supports matrix switching, integrates matrix functions within the processor, and supports single signal source to display multiple windows at the same time.

Rich interface, input support HDMI/VGA/DVI/SDI, output support HDMI/DVI.

Video Image Processing

◆ Internet of Things Big Data Decision Application Software Platform

The big data platform software of the Internet of Things uses in-depth learning and large data processing technology, and splices up the wall display in different regions to provide users with real-time monitoring status data of sensors. It also automatically monitors emergencies and decision support according to the level of early warning and linkage video.

Application Software for Big Data Decision Making in Internet of Things

5. Industry Solutions

1. Wisdom Community

—— Intelligent Property, Intelligent Well Cover, Intelligent Street Lamp, Intelligent Meter, Intelligent Water Meter, Intelligent River, Bridge Internet of Things

Intelligent meters, intelligent covers, intelligent rivers, etc. use NB-IOT + video surveillance convergence upload terminal, through the cloud, to achieve large data display and decision-making.

Application Scenario of Intelligent Community Ecosphere

Intelligent Cover Application Scenario

Smart Street Lamp Application Scene

1. Smart site

—— Municipal Engineering, Tunnel Engineering, Bridge Engineering

Smart site

3. Business Chain

—— Chain Brand Stores, Chain Express Points, Chain Supermarkets, Chain Shops, Chain Hotels, Chain Kindergarten Education

4. Snow Bright Project

—— Ping'an City, Ping'an Village

5. Intelligent Agriculture

—— Greenhouse planting, aquaculture, fruits and vegetables