CN107217718B - Feed robot and method based on image procossing - Google Patents

Abstract

The present invention relates to a kind of feed robot based on image procossing, the robot includes storage device, for being pre-placed different types of pipe-dredging agent；Starter is rinsed, for the water inside cistern to be released automatically；Average current detection means, the water velocity for the seat inner space to closestool are detected in real time, obtain real-time water velocity, and calculate the average current velocity in this flushing process；Type determination device, for determining the type for the pipe-dredging agent to be discharged based on average current velocity；Automatic releasing device, for certain type of pipe-dredging agent to be discharged into the seat inner space of closestool to realize the dredging of pipeline in the seat inner space to closestool.The invention further relates to a kind of material supply method based on image procossing.By means of the invention it is possible to realize the intelligent supply of dredging material, provided conveniently for the use of closestool user.

Description

Feeding robot and method based on image processing

The present invention is a divisional application of a patent with the application number 201710091418.1, the application date is February 21, 2017, and the invention name is "a feeding robot and method based on image processing".

technical field

The invention relates to the field of image processing, in particular to a feeding robot and method based on image processing.

Background technique

The first detailed written record of the toilet is the "wooden horse" in Ouyang Xiu's "Return to Field Record II" in the Northern Song Dynasty of China, which is explained as "wooden toilet" in "Ciyuan". The toilet used by the folks in ancient China is a round wooden barrel with a cover, which is painted with tung oil or good waterproof vermilion lacquer.

In 1596, English nobleman John Harrington invented the first practical toilet—a wooden seat with a cistern and flush valve.

In 1778, the British inventor Joseph Brame improved the design of the flush toilet, using a three-ball valve to control the flow of water in the tank, and a U-shaped elbow.

In the 19th century, the British government enacted a law requiring that every house must be equipped with a proper sewage treatment robot, and toilets began to improve significantly.

In 1861, Thomas Clapper, a British plumber, invented an advanced water-saving flushing robot, and waste discharge began to enter the modern era.

In 1885, Thomas Tuweifu obtained the first patent for an all-ceramic toilet in the UK, and dozens of improved patents were granted every year thereafter. In 1914, the Qixin Ceramics Factory (the predecessor of Tangshan Ceramics Factory) opened by the British in Tangshan manufactured the first ceramic toilet in China.

Flush toilets became popular in Europe and the United States in the 1860s, and later spread to Japan, South Korea and other Asian countries.

It can be seen that the toilet has a long history and its basic structure has been finalized for a long time. However, people still encounter the following problems when using it: when the toilet is blocked, the degree of blockage inside the toilet and the blocking material are not sure. Therefore, when dredging , had to choose one by one to carry out manual processing, even if it is so, it may not be able to achieve the dredging effect.

Contents of the invention

In order to solve the above problems, the present invention provides a feeding robot and method based on image processing. First, various dredging materials with different dredging strengths are pre-stored in different accommodation spaces on the toilet body. Secondly, the average water flow velocity and The type of target blockage is the determining factor of the blockage state. Again, high-precision sensors and image recognition equipment are used to accurately detect the above elements, and the required dredging materials are determined according to the detection results, and the accommodation space where the required dredging materials are located is automatically opened, so as to realize Automatic unclogging of the toilet.

According to an aspect of the present invention, a kind of feeding robot based on image processing is provided, and described robot comprises:

The storage device is used to pre-place different types of drain cleaners with different chemical compositions in different accommodation spaces on the toilet;

The flushing activation device is used to automatically discharge the water inside the toilet tank to flush the interior space of the toilet seat;

The average water flow detection device is used to detect the water flow velocity in the inner space of the toilet seat in real time, obtain the real-time water flow velocity, and calculate the average water flow velocity in this flushing process based on the real-time water flow velocity;

A type determination device, connected to the average water flow detection device, for determining the type of the pipe dredging agent to be released based on the average water flow velocity;

An automatic release device, connected with the type determination device, is used to release a certain type of pipe dredging agent into the seat interior space of the toilet to realize the dredging of the pipe in the seat interior space of the toilet;

Among them, various types of pipe dredging agents include: sodium hydroxide, sodium dichloroisocyanurate, sodium hypochlorite, tartaric acid, pull-off powder BX-78, bentonite, and essence. The different chemical compositions of pipe dredging agents lie in sodium hydroxide Different types of pipe dredging agents are divided into weak pipe dredging agents, general pipe dredging agents, strong pipe dredging agents and fast pipe dredging agents based on the percentage of ingredients of sodium hydroxide.

More specifically, in the image processing-based feeding robot, the type determination device specifically includes:

The data collection unit is used for collecting high-definition image data of the seat interior space of the toilet to obtain high-definition digital images;

The first filtering unit is connected to the data acquisition unit, and is used to perform edge-preserving filtering on the high-definition digital image to remove salt and pepper noise or impulse noise in the high-definition digital image, and obtain a first filtered image;

The second filtering unit, connected to the first filtering unit, is used to perform Gaussian filtering processing on the first filtered image to remove Gaussian noise in the first filtered image, and obtain a second filtered image, and the Gaussian noise includes a maximum amplitude value Large-amplitude Gaussian noise exceeding the preset amplitude value and small-amplitude Gaussian noise whose maximum amplitude value is less than or equal to the preset amplitude value;

A binarization unit, connected to the second filtering unit, for binarizing the second filtered image to obtain a binarized image;

The target matching unit is connected with the binarization unit, and is used to perform target matching processing on the binarized image to determine the type of blockage in the binarized image and output it as the type of target blockage. The target matching process includes binary The valued image is matched with the reference patterns of various types of blockages one by one to determine the type of blockage in the binarized image;

The type determination unit is respectively connected with the average water flow detection device and the target matching unit, and is used to determine the type of the pipeline dredging agent to be released based on the average water flow velocity and the target blockage type, wherein the smaller the average water flow velocity, the more certain the type of pipeline dredge The greater the proportion of sodium hydroxide in the agent, the greater the difficulty of corrosion corresponding to the target blockage type. The greater the proportion of sodium hydroxide in the determined type of pipeline dredging agent, when the average water flow rate is greater than the preset water flow threshold, it is determined The type of pipe dredge is none;

Wherein, said performing Gaussian filtering on the first filtered image to remove Gaussian noise in the first filtered image includes: performing adaptive recursive filtering on the first filtered image to remove small-amplitude Gaussian noise to obtain an intermediate filtered image, The intermediate filtered image is subjected to Wiener filtering to remove large-magnitude Gaussian noise to obtain a second filtered image;

Among them, various types of blockages include oil stains, vegetable leaves, tea leaves, hair, toilet paper, feces, kitchen waste, plastic bags and cloth.

More specifically, in the image processing-based feeding robot, the automatic release device specifically includes:

a switch unit for opening the opening of the containing space where the certain type of pipe unblocking agent is located to automatically release the certain type of pipe unblocking agent;

Wherein, when the determined type of pipe dredging agent is absent, the switch unit does not open the opening of the containing space where any type of pipe dredging agent is located, and sends a manual flushing signal.

More specifically, in the feeding robot based on image processing, it also includes:

A flushing number determination device, connected to the automatic release device, is used to determine the number of manual flushings based on the average water flow rate;

The display device is respectively connected with the automatic release device and the flushing times determination device, and is used for displaying the manual flushing signal and the number of times of manual flushing, so as to remind the toilet user.

More specifically, in the feeding robot based on image processing, it also includes:

A dose detection device, used to detect the current dose of the pipeline dredging agent in each accommodation space;

The dose alarm device is connected with the dose detection device, and is used to send out an insufficient dose alarm signal when the current dose of a certain type of pipe dredging agent is less than the preset dose, so as to remind the toilet user to add a corresponding type of pipe dredging agent.

According to another aspect of the present invention, there is provided a material supply method based on image processing, the method comprising:

Pre-place different types of drain cleaners with different chemical compositions in different accommodation spaces on the toilet;

Automatically discharge the water inside the toilet tank to flush the internal space of the toilet seat;

Real-time detection of the water flow velocity in the inner space of the toilet seat to obtain the real-time water flow velocity, and calculate the average water flow velocity in this flushing process based on the real-time water flow velocity;

determining the type of pipe cleaner to be released based on the average water velocity;

releasing a drain-cleaning agent of a determined type into the seat interior of the toilet to effect unclogging of the pipes in the seat interior of the toilet;

Among them, various types of pipe dredging agents include: sodium hydroxide, sodium dichloroisocyanurate, sodium hypochlorite, tartaric acid, pull-off powder BX-78, bentonite, and essence. The different chemical compositions of pipe dredging agents lie in sodium hydroxide Different types of pipe dredging agents are divided into weak pipe dredging agents, general pipe dredging agents, strong pipe dredging agents and fast pipe dredging agents based on the percentage of ingredients of sodium hydroxide.

More specifically, in the image processing-based material supply method, the determination of the type of pipe clearing agent to be released based on the average water flow velocity specifically includes the following steps:

Collecting high-definition image data on the interior space of the toilet seat to obtain a high-definition digital image; performing edge-preserving filtering on the high-definition digital image to remove salt and pepper noise or impulse noise in the high-definition digital image, and obtaining a first filtered image;

Performing Gaussian filtering on the first filtered image to remove Gaussian noise in the first filtered image to obtain a second filtered image, the Gaussian noise includes Gaussian noise with a large amplitude value exceeding a preset amplitude value and a maximum amplitude value less than Gaussian noise with a small amplitude value equal to a preset amplitude value; performing binarization processing on the second filtered image to obtain a binarized image;

Perform target matching processing on the binarized image to determine the type of blockage in the binarized image and output it as the type of target blockage, the target matching process includes comparing the binary image with the reference patterns of various types of blockages Perform shape matching to determine the type of blockage in the binarized image;

Determine the type of pipe unblocking agent to be released based on the average water flow velocity and the target block type, wherein the smaller the average water flow velocity, the greater the proportion of sodium hydroxide in the determined type of pipe unblocking agent, and the corrosion rate corresponding to the target block type The greater the difficulty, the greater the proportion of sodium hydroxide in the determined type of pipe dredging agent. When the average water flow velocity is greater than the preset water flow threshold, the determined type of pipe dredging agent is none;

Wherein, said performing Gaussian filtering on the first filtered image to remove Gaussian noise in the first filtered image includes: performing adaptive recursive filtering on the first filtered image to remove small-amplitude Gaussian noise to obtain an intermediate filtered image, The intermediate filtered image is subjected to Wiener filtering to remove large-magnitude Gaussian noise to obtain a second filtered image;

Among them, various types of blockages include oil stains, vegetable leaves, tea leaves, hair, toilet paper, feces, kitchen waste, plastic bags and cloth.

More specifically, in the image processing-based material supply method, the release of a certain type of plumbing agent into the seat interior space of the toilet so as to realize the dredging of the pipes in the seat interior space of the toilet specifically includes the following steps:

opening the opening of the containment space in which the defined type of pipe clearing agent is located to automatically release the determined type of pipe clearing agent;

Wherein, when it is determined that there is no type of pipe unblocking agent, the opening of the containing space where any type of pipe unblocking agent is located is not opened, and a manual flushing signal is sent.

More specifically, in the image processing-based material supply method, after the release of the determined type of plumbing agent into the seat interior space of the toilet to realize the dredging of the pipes in the seat interior space of the toilet, the method Also includes the following steps:

The number of manual flushing is determined based on the average water flow rate; while the manual flushing signal is displayed, the number of manual flushing is displayed to remind the toilet user.

More specifically, in the material supply method based on image processing, the following steps are further included: detecting the current dosage of the pipe dredging agent in each containing space;

When the current dose of a certain type of pipe dredging agent is less than the preset dose, an insufficient dose alarm signal is sent to remind the toilet user to add a corresponding type of pipe dredging agent.

In summary, it can be concluded that the key invention points of the present invention are:

First of all, various dredging materials with different dredging strengths are pre-stored, and an alarm will be issued when the dredging materials are insufficient;

Secondly, accurately detect the clogging state inside the toilet, and select the appropriate dredging material based on the clogging state;

Thirdly, suitable dredging materials are released for automatic dredging, and the whole process does not require manual participation, which realizes the full automation of toilet dredging.

Description of drawings

Embodiments of the present invention will be described below in conjunction with the accompanying drawings, wherein:

Fig. 1 is a structural block diagram of a feeding robot based on image processing according to an embodiment of the present invention.

Fig. 2 is a flow chart showing the steps of a material supply method based on image processing according to an embodiment of the present invention.

Reference signs: 1 storage device; 2 flushing activation device; 3 average water flow detection device; 4 type determination device; 5 automatic release device.

Detailed ways

The implementation of the image processing-based feeding robot of the present invention will be described in detail below with reference to the accompanying drawings.

The appearance of the toilet brings a lot of convenience to people. However, when the toilet is blocked, these conveniences are all converted into troubles. People need to use their brains and manually carry out common sense of the plan one by one according to the limited blockage situation they observe.

Common dredging options include:

1. Go to the grocery store to buy a manual dredge (something that looks like a spring), and manually dredge it;

2. Mop method: first put the mop on the mouth of the toilet, and then turn on the water to ensure that the water submerges the mop. Then press the mop hard, stuff it into the mouth of the toilet as much as possible, and finally lift the mop as fast as possible. Repeat this action, and the things in the toilet will be sucked out;

3. Find a longer plastic water pipe, connect one end to tap water, and plug the other end into the toilet mouth, then block the toilet mouth with a rag or something, and then turn on the tap water, which will use water pressure to clear the pipe;

4. If the toilet is blocked by hard objects, first use the toilet suction to remove the dirt, and put the water in while pouring water, so that there is only clean water in the toilet pool, then dry the water with a towel, and find a small place, such as makeup That kind of small mirror, a flashlight, put the small mirror at the bottom of the toilet, adjust the angle of the light source and the mirror to see a part of the inside of the pipe, usually small things will be stuck in the innermost part of the pipe that is about to turn downwards, find A more suitable iron wire is used to get stuck on small things to solve the problem.

5. Find a professional dredging company or dredging workers for manual dredging.

It can be seen that all dredging operations require manual observation, manual selection of dredging materials and dredging by hand. On the one hand, it is time-consuming and labor-intensive. On the other hand, it is easy to come into contact with the dirt during the dredging process. , and more importantly, if the scheme is wrong, the clogging problem cannot be solved, and the toilet cannot be used continuously, which will bring inconvenience to the life and work of the user.

In order to overcome the above deficiencies, the present invention builds a feeding robot and method based on image processing, adopts fully automatic dredging materials for correct selection of dredging materials and automatic leakage dredging, and at the same time determines that dredging materials are not needed. The number of manual flushing is required, and the user will be notified in time of the need for manual flushing and the number of flushing, so as to avoid the waste of dredging materials.

Fig. 1 is the structural block diagram of the feeding robot based on image processing shown according to the embodiment of the present invention, and described robot comprises:

The storage device is used to pre-place different types of drain cleaners with different chemical compositions in different accommodation spaces on the toilet;

The flushing activation device is used to automatically discharge the water inside the toilet tank to flush the interior space of the toilet seat;

The average water flow detection device is used to detect the water flow velocity in the inner space of the toilet seat in real time, obtain the real-time water flow velocity, and calculate the average water flow velocity in this flushing process based on the real-time water flow velocity;

A type determination device, connected to the average water flow detection device, for determining the type of the pipe dredging agent to be released based on the average water flow velocity;

An automatic release device, connected with the type determination device, is used to release a certain type of pipe dredging agent into the seat interior space of the toilet to realize the dredging of the pipe in the seat interior space of the toilet;

Among them, various types of pipe dredging agents include: sodium hydroxide, sodium dichloroisocyanurate, sodium hypochlorite, tartaric acid, pull-off powder BX-78, bentonite, and essence. The different chemical compositions of pipe dredging agents lie in sodium hydroxide Different types of pipe dredging agents are divided into weak pipe dredging agents, general pipe dredging agents, strong pipe dredging agents and fast pipe dredging agents based on the percentage of ingredients of sodium hydroxide.

Next, the specific structure of the feeding robot based on image processing of the present invention will be further described.

In the robot, the type determination device specifically includes:

The data collection unit is used for collecting high-definition image data of the seat interior space of the toilet to obtain high-definition digital images;

The first filtering unit is connected to the data acquisition unit, and is used to perform edge-preserving filtering on the high-definition digital image to remove salt and pepper noise or impulse noise in the high-definition digital image, and obtain a first filtered image;

The second filtering unit, connected to the first filtering unit, is used to perform Gaussian filtering processing on the first filtered image to remove Gaussian noise in the first filtered image, and obtain a second filtered image, and the Gaussian noise includes a maximum amplitude value Large-amplitude Gaussian noise exceeding the preset amplitude value and small-amplitude Gaussian noise whose maximum amplitude value is less than or equal to the preset amplitude value;

A binarization unit, connected to the second filtering unit, for binarizing the second filtered image to obtain a binarized image;

The target matching unit is connected with the binarization unit, and is used to perform target matching processing on the binarized image to determine the type of blockage in the binarized image and output it as the type of target blockage. The target matching process includes binary The valued image is matched with the reference patterns of various types of blockages one by one to determine the type of blockage in the binarized image;

The type determination unit is respectively connected with the average water flow detection device and the target matching unit, and is used to determine the type of the pipeline dredging agent to be released based on the average water flow velocity and the target blockage type, wherein the smaller the average water flow velocity, the more certain the type of pipeline dredge The greater the proportion of sodium hydroxide in the agent, the greater the difficulty of corrosion corresponding to the target blockage type. The greater the proportion of sodium hydroxide in the determined type of pipeline dredging agent, when the average water flow rate is greater than the preset water flow threshold, it is determined The type of pipe dredge is None;

Wherein, said performing Gaussian filtering on the first filtered image to remove Gaussian noise in the first filtered image includes: performing adaptive recursive filtering on the first filtered image to remove small-amplitude Gaussian noise to obtain an intermediate filtered image, The intermediate filtered image is subjected to Wiener filtering to remove large-magnitude Gaussian noise to obtain a second filtered image;

Among them, various types of blockages include oil stains, vegetable leaves, tea leaves, hair, toilet paper, feces, kitchen waste, plastic bags and cloth.

In the robot, the automatic release device specifically includes:

a switch unit for opening the opening of the containing space where the certain type of pipe unblocking agent is located to automatically release the certain type of pipe unblocking agent;

Wherein, when the determined type of pipe dredging agent is absent, the switch unit does not open the opening of the containing space where any type of pipe dredging agent is located, and sends a manual flushing signal.

The robot also includes:

A flushing number determination device, connected to the automatic release device, is used to determine the number of manual flushings based on the average water flow rate;

The display device is respectively connected with the automatic release device and the flushing times determination device, and is used for displaying the manual flushing signal and the number of times of manual flushing, so as to remind the toilet user.

The robot also includes:

A dose detection device, used to detect the current dose of the pipeline dredging agent in each accommodation space;

The dose alarm device is connected with the dose detection device, and is used to send out an insufficient dose alarm signal when the current dose of a certain type of pipe dredging agent is less than the preset dose, so as to remind the toilet user to add a corresponding type of pipe dredging agent.

Fig. 2 is a flow chart of the steps of an image processing-based material supply method according to an embodiment of the present invention, the method comprising:

Pre-place different types of drain cleaners with different chemical compositions in different accommodation spaces on the toilet;

Automatically discharge the water inside the toilet tank to flush the internal space of the toilet seat;

Real-time detection of the water flow velocity in the inner space of the toilet seat to obtain the real-time water flow velocity, and calculate the average water flow velocity in this flushing process based on the real-time water flow velocity;

determining the type of pipe cleaner to be released based on the average water velocity;

releasing a drain-cleaning agent of a determined type into the seat interior of the toilet to effect unclogging of the pipes in the seat interior of the toilet;

Among them, various types of pipe dredging agents include: sodium hydroxide, sodium dichloroisocyanurate, sodium hypochlorite, tartaric acid, pull-off powder BX-78, bentonite, and essence. The different chemical compositions of pipe dredging agents lie in sodium hydroxide Different types of pipe dredging agents are divided into weak pipe dredging agents, general pipe dredging agents, strong pipe dredging agents and fast pipe dredging agents based on the percentage of ingredients of sodium hydroxide.

Next, the specific flow of the material supply method based on image processing of the present invention will be further described.

In the method, the determination of the type of pipe dredging agent to be released based on the average water velocity specifically includes the following steps:

Collecting high-definition image data on the interior space of the toilet seat to obtain a high-definition digital image; performing edge-preserving filtering on the high-definition digital image to remove salt and pepper noise or impulse noise in the high-definition digital image, and obtaining a first filtered image;

Performing Gaussian filtering on the first filtered image to remove Gaussian noise in the first filtered image to obtain a second filtered image, the Gaussian noise includes Gaussian noise with a large amplitude value exceeding a preset amplitude value and a maximum amplitude value less than Gaussian noise with a small amplitude value equal to a preset amplitude value; performing binarization processing on the second filtered image to obtain a binarized image;

Perform target matching processing on the binarized image to determine the type of blockage in the binarized image and output it as the type of target blockage, the target matching process includes comparing the binary image with the reference patterns of various types of blockages Perform shape matching to determine the type of blockage in the binarized image;

Determine the type of pipe unblocking agent to be released based on the average water flow velocity and the target block type, wherein the smaller the average water flow velocity, the greater the proportion of sodium hydroxide in the determined type of pipe unblocking agent, and the corrosion rate corresponding to the target block type The greater the difficulty, the greater the proportion of sodium hydroxide in the determined type of pipe dredging agent. When the average water flow velocity is greater than the preset water flow threshold, the determined type of pipe dredging agent is none;

Wherein, said performing Gaussian filtering on the first filtered image to remove Gaussian noise in the first filtered image includes: performing adaptive recursive filtering on the first filtered image to remove small-amplitude Gaussian noise to obtain an intermediate filtered image, The intermediate filtered image is subjected to Wiener filtering to remove large-magnitude Gaussian noise to obtain a second filtered image;

Among them, various types of blockages include oil stains, vegetable leaves, tea leaves, hair, toilet paper, feces, kitchen waste, plastic bags and cloth.

In the method, the release of a certain type of pipe dredging agent into the seat interior space of the toilet so as to realize the dredging of the pipes in the seat interior space of the toilet specifically includes the following steps:

opening the opening of the containment space in which the defined type of pipe clearing agent is located to automatically release the determined type of pipe clearing agent;

Wherein, when it is determined that there is no type of pipe unblocking agent, the opening of the containing space where any type of pipe unblocking agent is located is not opened, and a manual flushing signal is sent.

In the method, after releasing the determined type of plumbing agent into the seat interior space of the toilet to realize the dredging of the pipes in the seat interior space of the toilet, the method further includes the following steps:

The number of manual flushing is determined based on the average water flow rate; while the manual flushing signal is displayed, the number of manual flushing is displayed to remind the toilet user.

In the method, the following steps are also included:

Detecting the current dosage of the pipe dredging agent in each containing space;

When the current dose of a certain type of pipe dredging agent is less than the preset dose, an insufficient dose alarm signal is sent to remind the toilet user to add a corresponding type of pipe dredging agent.

In addition, the robot may also include: a CDMA communication device, which is respectively connected to the automatic release device and the flushing times determination device, and is used to wirelessly send the manual flushing signal and the number of times of manual flushing to the mobile terminal held by the toilet user.

In addition, the method may also include the following steps:

The manual flushing signal and the number of times of manual flushing are wirelessly sent to the mobile terminal held by the toilet user.

Using the feeding robot and method based on image processing of the present invention, aiming at the technical problem of toilet dredging in the prior art that requires manual operation to participate in the whole process, through accurate detection of toilet clogging status, pre-storage of dredging materials, and customized selection mechanism of dredging materials And the response method when there is no need to dredge the material to complete the automatic dredging of the toilet. At the same time, it also performs a timely dose alarm when the dredging material is insufficient, thereby improving the intelligent level of the entire toilet.

It can be understood that although the present invention has been disclosed above with preferred embodiments, the above embodiments are not intended to limit the present invention. For any person skilled in the art, without departing from the scope of the technical solution of the present invention, the technical content disclosed above can be used to make many possible changes and modifications to the technical solution of the present invention, or be modified into equivalent changes, etc. effective example. Therefore, any simple modifications, equivalent changes and modifications made to the above embodiments according to the technical essence of the present invention, which do not deviate from the technical solution of the present invention, still fall within the protection scope of the technical solution of the present invention.

Claims (2)

1. a kind of feed robot based on image procossing, it is characterised in that the robot includes：

Storage device, for the different types of pipe-dredging agent of different chemical components to be placed into appearance different on closestool in advance Receive in space；

Starter is rinsed, is rushed for the water inside cistern to be released automatically with the seat inner space to closestool Wash；

Average current detection means, the water velocity for the seat inner space to closestool are detected in real time, are obtained real-time Water velocity, and the average current velocity in this flushing process is calculated based on real-time water velocity；

Type determination device, it is connected with the average current detection means, for determining what is discharged based on average current velocity The type of pipe-dredging agent；

Automatic releasing device, it is connected with the type determination device, for certain type of pipe-dredging agent to be discharged into closestool Seat inner space to realize the dredging of pipeline in the seat inner space to closestool；

Wherein, various types of pipe-dredging agents all include：Sodium hydroxide, sodium dichloro cyanurate, sodium hypochlorite, tartaric acid, Nekal BX-78, bentonite and essence, the different chemical components of pipe-dredging agent are that the dispensing percentage of sodium hydroxide is different, By dispensing percentage of various types of pipe-dredging agents based on sodium hydroxide number be divided into weak property pipe-dredging agent, general pipe Pipeline dredging agent, strong pipe dredging agent and fast siphunculus pipeline dredging agent；

The type determination device is specifically included with lower unit：

Data acquisition unit, for carrying out high-definition image data acquisition to the seat inner space of closestool to obtain high-definition digital figure Picture；

First filter unit, be connected with the data acquisition unit, for high-definition digital image carry out holding edge filter with The salt-pepper noise or impulsive noise in high-definition digital image are removed, obtains the first filtering image；

Second filter unit, be connected with first filter unit, for the first filtering image carry out gaussian filtering process with The Gaussian noise in the first filtering image is removed, obtains the second filtering image, the Gaussian noise exceedes including maximum amplitude value The significantly angle value Gaussian noise and maximum amplitude value of predetermined amplitude value are less than or equal to the amplitude value Gaussian noise of predetermined amplitude value；

Binarization unit, it is connected with the second filter unit, for carrying out binary conversion treatment to the second filtering image to obtain two-value Change image；

Object matching unit, is connected with binarization unit, is handled for carrying out object matching to binary image to determine two-value Change the type of the tamper in image and include as the output of target tamper type, the object matching processing by binary picture As the reference pattern with all kinds tamper carries out form fit to determine the type of the tamper in binary image one by one；

Type determining units, it is connected respectively with average current detection means and object matching unit, for based on average water flow velocity Degree and target tamper type determine the type for the pipe-dredging agent to be discharged, wherein, average current velocity is smaller, determines type Pipe-dredging agent in sodium hydroxide dispensing percentage it is bigger, corrosion difficulty is bigger corresponding to target tamper type, it is determined that The dispensing percentage of sodium hydroxide is bigger in the pipe-dredging agent of type, when average current velocity is more than default current threshold value, Certain type of pipe-dredging agent is nothing；

Wherein, it is described that gaussian filtering process is carried out to the first filtering image to remove the Gaussian noise bag in the first filtering image Include：Adaptive recursive filtering processing is carried out to the first filtering image to remove amplitude value Gaussian noise to obtain median filter figure Picture, Wiener filtering processing is carried out to middle filtering image with remove significantly angle value Gaussian noise to obtain the second filtering image；

Wherein, all kinds tamper include greasy dirt, dish leaf, tealeaves, hair, toilet paper, excrement, rubbish from cooking, polybag and Cloth；

The automatic releasing device is specifically included with lower unit：

Switch element, the opening for receiving space where opening certain type of pipe-dredging agent determine type with automatic release Pipe-dredging agent；

Wherein, when certain type of pipe-dredging agent be without when, switch element is not opened where any type of pipe-dredging agent The opening of receiving space, send artificial flushing signal.

2. a kind of material supply method based on image procossing, it is characterised in that methods described includes：

The different types of pipe-dredging agent of different chemical components is placed into receiving space different on closestool in advance；

Water inside cistern is released automatically and is rinsed with the seat inner space to closestool；

The water velocity of the seat inner space of closestool is detected in real time, obtains real-time water velocity, and be based on Real-time Water Flow velocity degree calculates the average current velocity in this flushing process；

The type for the pipe-dredging agent to be discharged is determined based on average current velocity；

Certain type of pipe-dredging agent is discharged into the seat inner space of closestool to realize the seat inner space to closestool The dredging of middle pipeline；

Wherein, various types of pipe-dredging agents all include：Sodium hydroxide, sodium dichloro cyanurate, sodium hypochlorite, tartaric acid, Nekal BX-78, bentonite and essence, the different chemical components of pipe-dredging agent are that the dispensing percentage of sodium hydroxide is different, By dispensing percentage of various types of pipe-dredging agents based on sodium hydroxide number be divided into weak property pipe-dredging agent, general pipe Pipeline dredging agent, strong pipe dredging agent and fast siphunculus pipeline dredging agent；

It is described to determine that the type for the pipe-dredging agent to be discharged specifically includes following steps based on average current velocity：

High-definition image data acquisition is carried out to obtain high-definition digital image to the seat inner space of closestool；

Holding edge filter is carried out to high-definition digital image to remove the salt-pepper noise or impulsive noise in high-definition digital image, is obtained Obtain the first filtering image；

Gaussian filtering process is carried out to the first filtering image to remove the Gaussian noise in the first filtering image, obtains the second filtering Image, it is small more than the significantly angle value Gaussian noise and maximum amplitude value of predetermined amplitude value that the Gaussian noise includes maximum amplitude value In the amplitude value Gaussian noise equal to predetermined amplitude value；

Binary conversion treatment is carried out to the second filtering image to obtain binary image；

Object matching processing is carried out to binary image to determine the type of the tamper in binary image and be blocked up as target Plug thing type exports, and the object matching processing includes one by one entering the reference pattern of binary image and all kinds tamper Row form fit is to determine the type of the tamper in binary image；

The type for the pipe-dredging agent to be discharged is determined based on average current velocity and target tamper type, wherein, average water Flow velocity degree is smaller, and the dispensing percentage of sodium hydroxide is bigger in certain type of pipe-dredging agent, and target tamper type is corresponding Corrosion difficulty it is bigger, the dispensing percentage of sodium hydroxide is bigger in certain type of pipe-dredging agent, works as average current velocity During more than default current threshold value, certain type of pipe-dredging agent is nothing；

Wherein, it is described that gaussian filtering process is carried out to the first filtering image to remove the Gaussian noise bag in the first filtering image Include：Adaptive recursive filtering processing is carried out to the first filtering image to remove amplitude value Gaussian noise to obtain median filter figure Picture, Wiener filtering processing is carried out to middle filtering image with remove significantly angle value Gaussian noise to obtain the second filtering image；

Wherein, all kinds tamper include greasy dirt, dish leaf, tealeaves, hair, toilet paper, excrement, rubbish from cooking, polybag and Cloth；

It is described that certain type of pipe-dredging agent is discharged into the seat inner space of closestool to realize inside the seat to closestool The dredging of pipeline specifically includes following steps in space：

The opening of receiving space where opening certain type of pipe-dredging agent to discharge certain type of pipe-dredging agent automatically；

Wherein, when certain type of pipe-dredging agent be without when, do not open receiving space where any type of pipe-dredging agent Opening, send artificial flushing signal.