CN210573854U

CN210573854U - Low-power-consumption image video real-time recognition device

Info

Publication number: CN210573854U
Application number: CN201922019715.2U
Authority: CN
Inventors: 张树琪
Original assignee: Zhuhai Jinrui Electric Power Technology Co ltd
Current assignee: Zhuhai Jinrui Electric Power Technology Co ltd
Priority date: 2019-11-21
Filing date: 2019-11-21
Publication date: 2020-05-19
Anticipated expiration: 2029-11-21

Abstract

The utility model aims at providing a long service life, the fast low-power consumption image video real-time identification equipment of reaction speed. The utility model discloses an identification module and power module, power module includes solar panel, super capacitor, battery and charge-discharge controller, solar panel super capacitor and the equal electric connection of battery in charge-discharge controller, power module does the identification module power supply. The utility model discloses be applied to open-air supervisory equipment's technical field.

Description

Low-power-consumption image video real-time recognition device

Technical Field

The utility model discloses be applied to field monitoring equipment's technical field, in particular to low-power consumption image video real-time identification equipment.

Background

The security monitoring under the ordinary scene generally adopts active power supply, does not need to consider the power consumption problem, but under the field condition, generally adopts passive power supply, uses solar energy to combine the battery to realize the power supply, is subject to the restriction of solar energy power generation and battery technique, and field equipment is extremely sensitive to the power consumption, if equipment power consumption is great, then has to adopt intermittent operation mode, can not continuous work, simultaneously because battery charge-discharge number of times is too much, leads to equipment life-span to shorten. In addition, in the prior art, a mode of video storage at the front end is generally adopted, and the images are taken regularly and transmitted back to the server for intensive operation. Therefore, the existing field equipment does not provide an intelligent detection function, only provides basic video recording and face detection functions, and at the moment, if the identification function is to be realized, videos need to be transmitted back to a server for processing, so that a large amount of electric quantity and flow can be consumed. Intensive operation is carried out at the server side, and when the number of devices is large, the bandwidth and the computing resources required by the server side are very large, such as a central processing unit and a graphic processor; and because of the intensive operation, the processing time may become long, and real-time processing cannot be provided. There is a chinese patent of publication number 209419738U at present, and it discloses a field monitoring device who facilitates use, and it has the rotation through wind-force automatically regulated solar cell panel, reduces the area of facing the wind, can resist the effect of strong wind weather, and it adopts solar cell panel and battery cooperation as power supply unit, can continuously charge and discharge the battery during the use, leads to the life of battery to shorten.

SUMMERY OF THE UTILITY MODEL

The utility model aims to solve the technical problem that overcome prior art not enough, provide a long service life, the fast low-power consumption image video real-time identification equipment of reaction speed.

The utility model adopts the technical proposal that: the utility model discloses an identification module and power module, power module includes solar panel, super capacitor, battery and charge-discharge controller, solar panel super capacitor and the equal electric connection of battery in charge-discharge controller, power module does the identification module power supply.

By the scheme, the charging and discharging times of the super capacitor can reach 10 ten thousand times, the capacity can not be reduced due to the fact that the memory effect is avoided, the charging speed is high, the super capacitor can be charged with electric quantity quickly and can be supplied with power to a load as long as the solar panel can provide a small amount of electric quantity, the service life of the low-power-consumption image video real-time identification device can be effectively prolonged by adopting the super capacitor as a first power supply source, the storage battery is used as a standby power supply, the super capacitor is used as an emergency power supply when the electric quantity of the super capacitor is consumed up, and therefore the identification module can continuously record videos under the condition of sufficient illumination intensity, and can also.

One preferred scheme is that the identification module comprises a camera, a processor, a memory and a mobile communication module, and the camera, the memory and the mobile communication module are all electrically connected to the processor.

According to the scheme, the camera is used for shooting images and videos, the processor is used for processing data and comparing object characteristics, the memory is used for storing image files, video files and object files, the object files are image files of specific faces, objects and the like, the images shot by the processor are compared with the object files, front-end identification is achieved, the effect of feeding back the images to the server after quick identification is achieved, meanwhile, network delay and processing delay caused by the fact that a traditional intelligent detection method can be identified only by the server processing are avoided, and quick response is achieved.

Preferably, the charge and discharge controller comprises a solar charging chip with model number BQ 24650.

Preferably, the memory is a memory with the model number of H26M41204 HPR.

Drawings

Fig. 1 is a schematic structural diagram of the present invention;

FIG. 2 is a flow chart of the present invention;

FIG. 3 is a circuit configuration diagram of the charge and discharge controller;

FIG. 4 is a circuit connection diagram of the identification module;

FIG. 5 is a schematic circuit diagram of a first portion of the identification module;

FIG. 6 is a schematic circuit diagram of a second portion of the identification module;

FIG. 7 is a schematic circuit diagram of a third portion of the identification module;

FIG. 8 is a schematic circuit diagram of a fourth portion of the identification module;

FIG. 9 is a schematic circuit diagram of a fifth portion of the identification module;

FIG. 10 is a schematic circuit diagram of a sixth portion of the identification module;

FIG. 11 is a schematic circuit diagram of a seventh portion of the identification module;

fig. 12 is a schematic circuit diagram of an eighth part of the identification module;

fig. 13 is a schematic circuit diagram of a ninth part of the identification module.

Detailed Description

As shown in fig. 1 to 13, in this embodiment, the utility model discloses an identification module and power module, power module includes solar panel 1, super capacitor 2, battery 3 and charge-discharge controller 4, solar panel 1 super capacitor 2 and the equal electric connection of battery 3 in charge-discharge controller 4, power module does the identification module power supply. The identification module comprises a camera 5, a processor 6, a memory 7 and a mobile communication module 8, wherein the camera 5, the memory 7 and the mobile communication module 8 are electrically connected to the processor 6.

In this embodiment, the charge and discharge controller 4 includes a solar charging chip of type BQ24650, the charge and discharge controller is an existing solar charging controller, an internal circuit of the charge and discharge controller includes a filtering circuit, an anti-reverse connection circuit, a battery switch circuit and a solar energy conversion portion, the filtering circuit is used for filtering harmful signals such as lightning stroke, surge and static electricity, the anti-reverse connection circuit is used for preventing the positive electrode and the negative electrode of the solar energy from being connected inversely to damage an electronic circuit, the battery switch circuit is used for controlling a battery switch, and the solar energy conversion portion converts the solar energy into electric energy of 12V to supply a stable direct current for charging the storage battery 3 and the super capacitor 2.

In this embodiment, the processor 6 includes an ATLAS200 AI acceleration module, which is a high-performance AI intelligent computation module, integrates an ascii Ascend 310 AI processor, can implement analysis and inference computation of various data such as images and videos, and can be widely used in scenes such as intelligent monitoring, robots, unmanned aerial vehicles, video servers, and the like. The memory 7 comprises a memory with the model number of H26M41204HPR and an EEPROM memory chip circuit, the model number of the EEPROM memory chip is 24LC256, the EEPROM memory chip is used for reading, writing and storing data by the ATLAS200 acceleration module through an I2C interface, the required voltage for the work of the 24LC256 chip is DC3.3V, the level of the I2C interface of the ATLAS200 acceleration module is 1.8V, and the conversion work is completed through a level conversion circuit. The ATLAS200 acceleration module has two startup modes, the first is TF card startup mode, the second is EMMC startup mode, in this embodiment, TF card startup mode is the main startup mode, and EMMC startup mode is the alternative. The ESD circuit is used for electrostatic discharge of the EMMC circuit. The TF card is made into a starting disk through external software, the starting disk is connected into a TF card interface circuit, and the ATLAS200 acceleration module reads starting information in the TF card through an SDIO interface to complete system starting. The UART communication interface circuit is used for debugging hardware peripherals, is connected with an external computer through the circuit, and debugs problems occurring in the running of equipment through serial port printing information. Because the ATLAS200 acceleration module does not have enough GPIO interfaces, the GPIO interface circuit is used for expanding the GPIO interfaces and is expanded through an I2C interface chip with the model of PCA6416AE, and the GPIO interfaces are used for controlling the starting state indicating circuit. The ATLAS200 acceleration module generates heat due to power consumption of a chip, heat is dissipated through an external fan, and a cooling fan driving interface circuit is used for power supply and rotating speed control of the external fan. The RESET circuit is used for carrying out normal starting and low-voltage detection of the ATLAS200 acceleration module, and a high-precision RESET chip with the model number of ADM6823 is adopted. The ATLAS200 acceleration module is integrated with an RGMII controller, a GE-PHY interface circuit is required to convert signals outside, and the signal is connected with an RJ45 interface after conversion, wherein the type of a PHY chip is 88E 1512. The ATLAS200 acceleration module has only one power input, signal name VSYS, and uses an input voltage with a typical value of 3.8V, and the transient current is close to 8A under the maximum power consumption. In addition, power supply circuits DC3.3V and DC1.8V are used for circuits such as EEPROM memory chip circuits of the backplane.

The work flow of the identification module is as follows: when the camera 5 shoots, data are output and input into the ATLAS200 acceleration module after passing through the internet access circuit and the GE-PHY interface circuit, the ATLAS200 acceleration module carries out reasoning on data decoding according to a preset program, the ATLAS200 acceleration module outputs the data to the mobile communication module 8 when an object is detected, and the mobile communication module 8 uploads the data to a server.

As shown in fig. 2, the work flow of the present invention includes the following steps:

A. the power supply module supplies power to the identification module by adopting a timely switching method, and the identification module reads the electric quantity information of the super capacitor 2 and the illuminance received by the solar power generation panel 1;

B. under the environment with low illuminance, the identification module adopts a long-interval timing shooting mode, the camera 5 starts to shoot the surrounding environment at intervals and stores image information in the memory 7, the processor 6 compares the currently shot image with the last shot image, the timing shooting mode is kept when the difference degree is low, the video shooting mode is activated when the difference degree is high, the camera 5 continuously carries out real-time video shooting on the difference position, meanwhile, the processor 6 compares the images to carry out object identification, when an object is detected, an instruction is sent out to enable the camera 5 to follow the shot video, the mobile communication module 8 is started to feed back alarm information to the server, and the video is uploaded to the server;

C. when the environment is sufficiently illuminated and the electric quantity of the super capacitor 2 is sufficient, the camera 5 starts continuous real-time shooting, the processor 6 identifies an object, and when the object is detected, an instruction is sent out to enable the camera 5 to follow the shooting, the mobile communication module 8 is started to feed back alarm information to the server, and a video is uploaded to the server;

D. in the environment with sufficient illumination and when the electric quantity of the super capacitor 2 is lower, the identification module adopts a short-interval timing shooting mode, the processor 6 compares the currently shot image with the last shot image, and the video shooting mode is activated when the degree of distinction is high.

The timely switching method is that the solar power generation panel 1 is started after receiving illumination and outputs electric energy to the charge and discharge controller 4, the charge and discharge controller 4 transmits the electric energy to the super capacitor 2 and the storage battery 3, the super capacitor 2 is a first output source during discharging, the storage battery 3 is a second output source, the super capacitor 2 is preferentially adopted to output the electric energy under normal illumination, and meanwhile, the solar power generation panel 1 continuously charges the super capacitor 2 so as to provide the electric energy for a load; and when the electric quantity of the super capacitor 2 is exhausted in a low-light environment, the charge and discharge controller 4 switches the output source to the storage battery 3.

Claims

1. A low-power consumption image video real-time identification device is characterized in that: it includes discernment module and power module, power module includes solar panel (1), super capacitor (2), battery (3) and charge-discharge controller (4), solar panel (1) super capacitor (2) and the equal electric connection of battery (3) in charge-discharge controller (4), power module does the discernment module power supply.

2. The low-power consumption image video real-time identification device according to claim 1, characterized in that: the identification module comprises a camera (5), a processor (6), a memory (7) and a mobile communication module (8), wherein the camera (5), the memory (7) and the mobile communication module (8) are electrically connected to the processor (6).

3. The low-power consumption image video real-time identification device according to claim 1, characterized in that: the charging and discharging controller (4) comprises a solar charging chip with the model number of BQ 24650.

4. The low-power consumption image video real-time identification device according to claim 2, characterized in that: the memory (7) comprises a memory of type H26M41204 HPR.