CN106570854A - Water state detection device and method and image processing device - Google Patents
Water state detection device and method and image processing device Download PDFInfo
- Publication number
- CN106570854A CN106570854A CN201510648508.7A CN201510648508A CN106570854A CN 106570854 A CN106570854 A CN 106570854A CN 201510648508 A CN201510648508 A CN 201510648508A CN 106570854 A CN106570854 A CN 106570854A
- Authority
- CN
- China
- Prior art keywords
- real
- block
- time
- sub
- unit
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20024—Filtering details
Landscapes
- Engineering & Computer Science (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The embodiment of the present invention provides a water state detection device and method and an image processing device. The detection method comprises the steps of dividing a real-time image of recording the water into a plurality of real-time sub-blocks; carrying out the directional filtering processing on each real-time sub-block separately, and obtaining the filtering results of the different directions of each real-time sub-block; extracting the statistics characteristics separately according to the filtering results; based on a pre-obtained classifier model and the statistics characteristics, carrying out the type identification on each real-time sub-block; gathering the types of the plurality of real-time sub-blocks of the real-time image; and determining the water state corresponding to the real-time image according to the statistics result. Therefore, not only sensors are not needed when the water state is detected, but also a system is simple, is low in cost and can detect the water state real-timely and accurately.
Description
Technical field
The present embodiments relate to technical field of image processing, more particularly to a kind of detection means of water state, method with
And image processing equipment.
Background technology
(such as wastewater treatment process) needs to detect the kinestate of water under many scenes.When in water tank
Water from it is static be changed into motion when or from motion be changed into static when, to these changes, to carry out fast reaction be necessary and important
's.
At present some systems detected to water state (including water sport state and water resting state) are occurred in that,
For example using level sensor, temperature sensor, chlorine measurement sensor etc..By these sensors, Ke Yi
Water state is detected to a certain extent.
It should be noted that the introduction of technical background is intended merely to above it is convenient technical scheme is carried out it is clear,
Complete explanation, and facilitate the understanding of those skilled in the art and illustrate.Can not be merely because these schemes be at this
Bright background section is set forth and thinks that above-mentioned technical proposal is known to those skilled in the art.
The content of the invention
But, inventor has found:Carrying out the system of water state detection at present needs to use sensor, system complex and into
This height;And testing result is often delayed, it is impossible to detect in real time and exactly to water state.
Embodiments provide a kind of detection means of water state, method and image processing equipment.Simple system
And low cost, water state can be detected in real time and exactly.
One side according to embodiments of the present invention, there is provided a kind of detection means of water state, the detection means includes:
In real time partition unit, by the real time imaging of record water multiple real-time sub-blocks are divided into;
Real-Time Filtering unit, for each real-time sub-block distinguishes travel direction Filtering Processing, obtains that described each is real-time
The filter result of the different directions of sub-block;
Real-time characteristic extracting unit, according to the filter result of the different directions statistical nature is extracted respectively;
Type identification unit, based on the sorter model being obtained ahead of time and the statistical nature to described each real-time son
Block carries out respectively type identification;
Type statistics unit, counts to the type of the plurality of real-time sub-block of the real time imaging;And
Status determining unit, according to statistical result the corresponding water state of the real time imaging is determined.
Second aspect according to embodiments of the present invention, there is provided a kind of detection method of water state, the detection method bag
Include:
The real time imaging of record water is divided into into multiple real-time sub-blocks;
For each real-time sub-block distinguishes travel direction Filtering Processing, the different directions of each real-time sub-block are obtained
Filter result;
Statistical nature is extracted respectively according to the filter result of the different directions;
Class is carried out respectively to described each real-time sub-block based on the sorter model being obtained ahead of time and the statistical nature
Type is recognized;
The type of the plurality of real-time sub-block of the real time imaging is counted;And
The corresponding water state of the real time imaging is determined according to statistical result.
3rd aspect according to embodiments of the present invention, there is provided a kind of image processing equipment, including as above watery
The detection means of state.
The beneficial effect of the embodiment of the present invention is:Image division by record water is multiple sub-blocks;For each sub-block
Travel direction Filtering Processing and extract after statistical nature respectively, type identification is carried out respectively to each real-time sub-block;It is right
The type of multiple sub-blocks of described image is counted;And determine that described image is corresponding watery according to statistical result
State.Thus, sensor, simple system and low cost need not be not only used when water state is detected, and can be real
When ground water state is detected exactly.
With reference to explanation hereinafter and accompanying drawing, the particular implementation of the embodiment of the present invention is disclose in detail, specify this
The principle of bright embodiment can be in adopted mode.It should be understood that embodiments of the present invention in scope not thus
It is restricted.In the range of the spirit and terms of claims, embodiments of the present invention include many changes,
Modification and equivalent.
The feature for describing for a kind of embodiment and/or illustrating can be in same or similar mode one or more
It is combined with the feature in other embodiment used in individual other embodiment, or substitute in other embodiment
Feature.
It should be emphasized that term "comprises/comprising" refers to the presence of feature, one integral piece, step or component when using herein,
But it is not precluded from the presence of one or more further features, one integral piece, step or component or additional.
Description of the drawings
Included accompanying drawing is used for providing being further understood from the embodiment of the present invention, which constitutes of description
Point, for illustrating embodiments of the present invention, and come together to explain the principle of the present invention with word description.Obviously
Ground, drawings in the following description are only some embodiments of the present invention, for those of ordinary skill in the art,
Without having to pay creative labor, can be with according to these other accompanying drawings of accompanying drawings acquisition.In the accompanying drawings:
Fig. 1 is a schematic diagram of the detection method of the water state of the embodiment of the present invention 1;
Fig. 2 is another schematic diagram of the detection method of the water state of the embodiment of the present invention 1;
Fig. 3 is a schematic diagram of the directivity filtering of the embodiment of the present invention 1;
Fig. 4 is another schematic diagram of the directivity filtering of the embodiment of the present invention 1;
Fig. 5 is a schematic diagram of the detection means of the water state of the embodiment of the present invention 2;
Fig. 6 is another schematic diagram of the detection means of the water state of the embodiment of the present invention 2;
Fig. 7 is a composition schematic diagram of the image processing equipment of the embodiment of the present invention 3.
Specific embodiment
Referring to the drawings, by description below, the aforementioned and further feature of the embodiment of the present invention will be apparent from.
In the specification and illustrated in the drawings, only certain exemplary embodiments of this invention is specifically disclosed, which show and wherein can adopt this
The some embodiments of the principle of bright embodiment, it will thus be appreciated that the invention is not restricted to described embodiment, phase
Instead, the embodiment of the present invention includes whole modifications, modification and the equivalent for falling within the scope of the appended claims.
Embodiment 1
The embodiment of the present invention provides a kind of detection method of water state.Fig. 1 is the detection of the water state of the embodiment of the present invention
One schematic diagram of method, as shown in figure 1, the detection method includes:
Step 101, by the real time imaging of record water multiple real-time sub-blocks are divided into;
Step 102, for each real-time sub-block distinguishes travel direction Filtering Processing, obtains described each real-time sub-block
Different directions filter result;
Step 103, according to the filter result of the different directions of each real-time sub-block statistical nature is extracted respectively;
Step 104, based on the sorter model being obtained ahead of time and the statistical nature to described each real-time sub-block point
Type identification is not carried out;
Step 105, counts to the type of the plurality of real-time sub-block of the real time imaging;And
Step 106, according to statistical result the corresponding water state of the real time imaging is determined.
In the present embodiment, for the monitor area for needing to carry out water state detection, photographic head can be set and is carried out continuously
Shoot, obtain the video comprising multiframe real time imaging.For every frame real time imaging, can be according to above-mentioned steps 101
The corresponding water state of the two field picture is detected to step 106.
In the present embodiment, photographic head can be directed at the water surface and arrange or arrange at an angle.Wherein water state
Water sport state and water resting state can be included, water sport state for example can be the shape for blister (bubble) occur
There is the state of water wave in state, but the invention is not restricted to this, can also be other states.
In the present embodiment, sorter model can be obtained ahead of time, for example, the image of monitor area carried out in advance
Obtain after the training of a period of time.But the invention is not restricted to this, for example, can also previously generate based on experience value.
Fig. 2 is another schematic diagram of the detection method of the water state of the embodiment of the present invention, as shown in Fig. 2 the detection
Method can include training stage and monitor stages.
In the training stage, methods described includes:
Step 2011, by training image multiple training sub-blocks are divided into;
Step 2012, for each training sub-block difference travel direction Filtering Processing, obtains described each training
The filter result of the different directions of block;
Step 2013, according to the filter result of the different directions of each training sub-block statistical nature is extracted respectively;
Step 2014, by the statistical nature support vector machine (SVM) or artificial neural network (ANN) are input to
In obtaining sorter model.
In the specific implementation, it is possible to use photographic head obtains multiple training images, is iteratively performed above-mentioned steps 2011
To step 2014, it is possible thereby to obtain more accurately sorter model.
In monitor stages, methods described includes:
Step 2021, by the real time imaging of record water multiple real-time sub-blocks are divided into;
Step 2022, for each real-time sub-block distinguishes travel direction Filtering Processing, obtains described each real-time son
The filter result of the different directions of block;
Step 2023, according to the filter result of the different directions of each real-time sub-block statistical nature is extracted respectively;
Step 2024, based on the sorter model being obtained ahead of time and the statistical nature to described each real-time sub-block
Type identification is carried out respectively;
Step 2025, counts to the type of the plurality of real-time sub-block of the real time imaging;And
Step 2026, according to statistical result the corresponding water state of the real time imaging is determined.
It should be noted that above-mentioned training stage and monitor stages do not have ordering relation, training rank can be first carried out
Duan Ranhou performs monitor stages;Can also two stage pipelines carry out, it is possible thereby to constantly carry out to sorter model
Update and optimize.
In the present embodiment, image (real time imaging or training image) can be divided into multiple sub-blocks, for example, is divided
For the sub-block of 65*65, wherein unit is pixel;But the invention is not restricted to this, sub-block can be determined according to practical situation
Size or shape etc..
In step 2012 or step 2022, for the equal travel direction filtering of each sub-block, it is possible to use improved
Gabor filtering algorithms;But the invention is not restricted to this, for example can also be using other directivity filtering algorithms.Below
Illustrate by taking the method based on Gabor filtering as an example.
It is, for example possible to use equation below travel direction Filtering Processing:
Wherein, x'=xcos θ+ysin θ and y'=-xsin θ+ycos θ;X and y represent pixel coordinate, λ tables
Show the wavelength of the sine curve factor, θ represents benchmark to the direction of parallel band,Phase offset is represented, σ represents Gauss
Factor standard is poor, γ representation space aspect rates.
In the specific implementation, for example can be arranged as follows:
Wavelength:λ=10, in principle 2<λ<blocksize*0.2;Wherein blocksize is the size of sub-block;
Gauss factor standard is poor:σ=4;
Aspect rate:γ=1;
Phase offset:
It should be noted that schematically show only the value of each parameter above, but the invention is not restricted to this.
Wherein, for the yardstick in Gabor filtering, σ and λ is constituted into inversely prroportional relationship, scale has it to set
Formula.Typically indication is multiple dimensioned takes different center frequency (1/ λ).In the experiment of the embodiment of the present invention λ take 10 (can
It is interpreted as single yardstick);Graphical rule difference be not the present invention pay close attention to object, it is possible to take on yardstick
One general more common empirical value.
In the present embodiment, filtered using different θ values travel directions, to strengthen the difference between each sub-block;
Such as θ=45,90,135,180;
Fig. 3 is a schematic diagram of the directivity filtering of the embodiment of the present invention, is from left to right respectively illustrated with blister
Image, the filtered image of directivity that θ=90 are carried out to the image and the directivity of θ=180 is carried out to the image
Filtered image.
Fig. 4 is another schematic diagram of the directivity filtering of the embodiment of the present invention, is from left to right respectively illustrated with static
The image of the water surface, the filtered image of directivity that θ=90 are carried out to the image and θ=180 are carried out to the image
The filtered image of directivity.
As shown in Figures 3 and 4, filtered using different θ values travel directions, can be strengthened in water sport state
The difference of image and the image in water resting state.
In step 2013 or step 2023, can respectively be extracted according to the filter result of the different directions of each sub-block
Statistical nature.Wherein, statistical nature includes:Image first moment and image second order square.For example, statistical nature can be
Average and variance.
In the present embodiment, the filtered eigenvalue of directivity is not directly adopted, and uses statistical nature to each sub-block
Type identification is carried out respectively.
It is, for example possible to use equation below carries out the extraction of statistical nature:
Wherein, mean represents average, and var represents variance;M and N are respectively the real-time sub-block or the training
The height and width of sub-block, I () represents the pixel value of each pixel in the real-time sub-block or the training sub-block.
So, for each sub-block can obtain for example following characteristics of image:
F=[var1,mean1,var2,mean2,var3,mean3var4,mean4]
Wherein, subscript 1,2,3,4 represents respectively θ=45,90,135,180 situation.
Thus, the present embodiment utilization orientation filtering (such as Gabor filtering) enhancing to image direction, can be with
The filtered texture image of directivity is obtained to project the difference of kinetic water body and impounded body, it is possible thereby to reduce noise
Affect, and then texture image information is calculated.Additionally, the present embodiment is not directly using the feature of directivity filtering
Value, but adopt statistical nature, that is, the image pixel distribution arrangement difference that water sports are brought is only focused on, thus may be used
To reduce computation complexity, amount of calculation is reduced.
In step 2024, can be based on the sorter model that is obtained ahead of time and the statistical nature to it is described each
In real time sub-block carries out respectively type identification;
Specifically, for certain real-time sub-block, can be based on the sorter model and the real-time sub-block being obtained ahead of time
Statistical nature is scored and is obtained score value;In the case where the score value is more than predetermined threshold value (first threshold),
The real-time sub-block is defined as into type of sports;Otherwise the real-time sub-block is defined as into stationary kind.
With regard to specifically how to be scored, can be using any one methods of marking in prior art.
In step 2025, the type of the plurality of real-time sub-block of the real time imaging can be counted.By
This, for a certain image in all sub-blocks in any one, it may be determined that be type of sports or stationary kind.
In step 2026, the corresponding water state of the real time imaging can be determined according to statistical result.
Specifically, the number that the real-time sub-block of type of sports is confirmed as in the real time imaging is more than predetermined threshold value
In the case of (Second Threshold), the real time imaging is defined as into water sport state;It is otherwise that the real time imaging is true
It is set to water resting state.
For example, if a certain real time imaging includes 32 sub-blocks, when counting wherein 28 (more than Second Threshold 25)
When sub-block is type of sports, the real time imaging can be defined as water sport state.Wherein only have 12 when counting
When (being less than Second Threshold 25) sub-block is type of sports, the real time imaging can be defined as water resting state.
Table 1 diagrammatically illustrates the situation for determining water state.
Table 1
From above-described embodiment, the image division by record water is multiple sub-blocks;For each sub-block side of carrying out respectively
Tropism Filtering Processing is simultaneously extracted after statistical nature, and to each real-time sub-block type identification is carried out respectively;To described image
The type of multiple sub-blocks is counted;And the corresponding water state of described image is determined according to statistical result.Thus, exist
Sensor, simple system and low cost need not be not only used during detection water state, and can be right exactly in real time
Water state is detected.
Embodiment 2
The embodiment of the present invention provides a kind of detection means of water state, corresponding to the detection side of the water state in embodiment 1
Method.Content same as Example 1 is repeated no more.
Fig. 5 is a schematic diagram of the detection means of the water state of the embodiment of the present invention, as shown in figure 5, the detection dress
Putting 500 includes:
In real time partition unit 501, by the real time imaging of record water multiple real-time sub-blocks are divided into;
Real-Time Filtering unit 502, for each real-time sub-block distinguishes travel direction Filtering Processing, obtain it is described each
The filter result of the different directions of real-time sub-block;
Real-time characteristic extracting unit 503, extracts respectively according to the filter result of the different directions of each real-time sub-block
Statistical nature;
Type identification unit 504, based on the sorter model being obtained ahead of time and the statistical nature to described each reality
When sub-block carry out type identification respectively;
Type statistics unit 505, counts to the type of the plurality of real-time sub-block of the real time imaging;And
Status determining unit 506, according to statistical result the corresponding water state of the real time imaging is determined.
Fig. 6 is another schematic diagram of the detection means of the water state of the embodiment of the present invention, as shown in fig. 6, the detection
Device 600 includes each unit as above.
As shown in fig. 6, the detection means 600 also includes:
Training partition unit 601, by training image multiple training sub-blocks are divided into;
Training filter unit 602, for each training sub-block difference travel direction Filtering Processing, obtain it is described each
The filter result of the different directions of training sub-block;
Training characteristics extracting unit 603, extracts respectively according to the filter result of the different directions of each training sub-block
Statistical nature;And
Model signal generating unit 604, obtains during the statistical nature is input to into support vector machine or artificial neural network
The sorter model.
In the present embodiment, the Real-Time Filtering unit 502 or the training filter unit 602 can be using following public
Formula travel direction Filtering Processing:
Wherein, x'=xcos θ+ysin θ and y'=-xsin θ+ycos θ;X and y represent pixel coordinate, λ tables
Show the wavelength of the sine curve factor, θ represents benchmark to the direction of parallel band,Phase offset is represented, σ represents Gauss
Factor standard is poor, γ representation space aspect rates.
In the present embodiment, the Real-Time Filtering unit 502 or the training filter unit 602 can use different
θ value travel direction Filtering Processing is strengthening the difference between each sub-block;The statistical nature can include:Image single order
Square and image second order square.
In the present embodiment, the real-time characteristic extracting unit 503 or the training characteristics extracting unit 603 can make
The extraction of statistical nature is carried out with equation below:
Wherein, mean represents average, and var represents variance;M and N are respectively the real-time sub-block or the training
The height and width of sub-block, I () represents the pixel value of each pixel in the real-time sub-block or the training sub-block.
In the present embodiment, the type identification unit 504 can include:
Sub-block scoring unit, for certain real-time sub-block, based on the sorter model being obtained ahead of time and the real-time sub-block
Statistical nature scored and obtained score value;And
Type determining units, in the case where the score value is more than predetermined threshold value, by the real-time sub-block fortune are defined as
Dynamic type;Otherwise the real-time sub-block is defined as into stationary kind.
In the present embodiment, the status determining unit 506 specifically can be used for:It is confirmed as fortune in described image
Described image is defined as water sport state by the number of the real-time sub-block of dynamic type more than in the case of predetermined threshold value;It is no
Then described image is defined as into water resting state.
From above-described embodiment, the image division by record water is multiple sub-blocks;For each sub-block side of carrying out respectively
Tropism Filtering Processing is simultaneously extracted after statistical nature, and to each real-time sub-block type identification is carried out respectively;To described image
The type of multiple sub-blocks is counted;And the corresponding water state of described image is determined according to statistical result.Thus, exist
Sensor, simple system and low cost need not be not only used during detection water state, and can be right exactly in real time
Water state is detected.
Embodiment 3
The embodiment of the present invention provides a kind of image processing equipment, and described image processing equipment includes:As described in Example 2
Water state detection means.
Fig. 7 is a composition schematic diagram of the image processing equipment of the embodiment of the present invention.As shown in fig. 7, image procossing sets
Standby 700 can include:Central processing unit (CPU) 100 and memorizer 110;Memorizer 110 is coupled to centre
Reason device 100.Wherein the memorizer 110 can store various data;The program that additionally storage information is processed, and
The program is performed under the control of central processing unit 100.
In one embodiment, the function of the detection means 500 or 600 of water state can be integrated into central process
In device 100.Wherein, central processing unit 100 can be configured to realize the detection of water state as described in Example 1
Method.I.e. central processing unit 100 can be controlled as follows:
The real time imaging of record water is divided into into multiple real-time sub-blocks;For each real-time sub-block difference travel direction filter
Ripple process, obtains the filter result of the different directions of each real-time sub-block;According to described each real-time sub-block not
Equidirectional filter result extracts respectively statistical nature;Based on the sorter model being obtained ahead of time and the statistical nature
Type identification is carried out respectively to described each real-time sub-block;Type to the plurality of real-time sub-block of the real time imaging
Counted;And the corresponding water state of the real time imaging is determined according to statistical result.
In another embodiment, the detection means 500 or 600 of water state can separate with central processing unit 100
Configuration, for example, can be configured to the chip that is connected with central processing unit 100 by the detection means 500 or 600 of water state,
The function of the detection means 500 or 600 of water state is realized by the control of central processing unit.
Additionally, as shown in fig. 7, image processing equipment 700 can also include:Input-output unit 120 and display are single
Unit 130 etc.;Wherein, similarly to the prior art, here is omitted for the function of above-mentioned part.It should be noted that
Image processing equipment 700 is also not necessary to include all parts shown in Fig. 7;Additionally, image processing equipment
700 can also may be referred to prior art including the part being not shown in Fig. 7.
The embodiment of the present invention also provides a kind of computer-readable program, wherein when performing the journey in image processing equipment
During sequence, described program causes computer that the inspection of water state as described in Example 1 is performed in described image processing equipment
Survey method.
The embodiment of the present invention also provides a kind of storage medium of the computer-readable program that is stored with, wherein the computer can
Reader causes computer that the detection method of water state as described in Example 1 is performed in image processing equipment.
Apparatus and method more than of the invention can be realized by hardware, it is also possible to be realized by combination of hardware software.The present invention
It is related to such computer-readable program, when the program is performed by logical block, realizes can the logical block
Devices described above or component parts, or make the logical block realize various methods or step mentioned above.This
The bright storage medium further related to for storing procedure above, such as hard disk, disk, CD, DVD, flash memory
Deng.
Above in association with specific embodiment, invention has been described, it will be appreciated by those skilled in the art that this
A little descriptions are all exemplary, are not limiting the scope of the invention.Those skilled in the art can be according to this
Inventive principle makes various variants and modifications to the present invention, and these variants and modifications are also within the scope of the invention.
With regard to including the embodiment of above example, following note being also disclosed:
A kind of (note 1) detection means of water state, the detection means includes:
In real time partition unit, by the real time imaging of record water multiple real-time sub-blocks are divided into;
Real-Time Filtering unit, for each real-time sub-block distinguishes travel direction Filtering Processing, obtains that described each is real-time
The filter result of the different directions of sub-block;
Real-time characteristic extracting unit, according to the filter result of the different directions statistical nature is extracted respectively;
Type identification unit, based on the sorter model being obtained ahead of time and the statistical nature to described each real-time son
Block carries out respectively type identification;
Type statistics unit, counts to the type of the plurality of real-time sub-block of the real time imaging;And
Status determining unit, according to statistical result the corresponding water state of the real time imaging is determined.
The detection means of (note 2) according to note 1, wherein, the detection means also includes:
Training partition unit, by training image multiple training sub-blocks are divided into;
Training filter unit, for each training sub-block difference travel direction Filtering Processing, obtains described each training
The filter result of the different directions of sub-block;
Training characteristics extracting unit, according to the filter result of the different directions statistical nature is extracted respectively;And
Model signal generating unit, obtains described during the statistical nature is input to into support vector machine or artificial neural network
Sorter model.
The detection means of (note 3) according to note 1 or 2, wherein, the Real-Time Filtering unit or the instruction
Practice filter unit and use equation below travel direction Filtering Processing:
Wherein, x'=xcos θ+ysin θ and y'=-xsin θ+ycos θ;X and y represent pixel coordinate, λ tables
Show the wavelength of the sine curve factor, θ represents benchmark to the direction of parallel band,Phase offset is represented, σ represents Gauss
Factor standard is poor, γ representation space aspect rates.
The detection means of (note 4) according to note 3, wherein, the Real-Time Filtering unit or the training are filtered
Ripple unit strengthens the difference between each sub-block using different θ value travel direction Filtering Processing.
The detection means of (note 5) according to note 1 or 2, wherein, the statistical nature includes:Image one
Rank square and image second order square.
(note 6) detection means according to claim 5, wherein, the real-time characteristic extracting unit or institute
Stating training characteristics extracting unit carries out the extraction of statistical nature using equation below:
Wherein, mean represents average, and var represents variance;M and N are respectively the real-time sub-block or the training
The height and width of sub-block, I () represents the pixel value of each pixel in the real-time sub-block or the training sub-block.
The detection means of (note 7) according to note 1, wherein, the type identification unit includes:
Sub-block scoring unit, for certain real-time sub-block, based on the sorter model being obtained ahead of time and the real-time sub-block
Statistical nature scored and obtained score value;
Type determining units, in the case where the score value is more than predetermined threshold value, by the real-time sub-block fortune are defined as
Dynamic type;Otherwise the real-time sub-block is defined as into stationary kind.
The detection means of (note 8) according to note 1, wherein, the status determining unit specifically for:
It is confirmed as the number of real-time sub-block of type of sports in the real time imaging more than in the case of predetermined threshold value, will be described
Real time imaging is defined as water sport state;Otherwise the real time imaging is defined as into water resting state.
A kind of (note 9) detection method of water state, the detection method includes:
The real time imaging of record water is divided into into multiple real-time sub-blocks;
For each real-time sub-block distinguishes travel direction Filtering Processing, the different directions of each real-time sub-block are obtained
Filter result;
Statistical nature is extracted respectively according to the filter result of the different directions;
Class is carried out respectively to described each real-time sub-block based on the sorter model being obtained ahead of time and the statistical nature
Type is recognized;
The type of the plurality of real-time sub-block of the real time imaging is counted;And
The corresponding water state of the real time imaging is determined according to statistical result.
The detection method of (note 10) according to note 9, wherein, the detection method also includes:
Training image is divided into into multiple training sub-blocks;
For each training sub-block difference travel direction Filtering Processing, the different directions of each training sub-block are obtained
Filter result;
Statistical nature is extracted respectively according to the filter result of the different directions;And
The sorter model is obtained during the statistical nature is input to into support vector machine or artificial neural network.
The detection method of (note 11) according to note 9 or 10, wherein, using equation below travel direction
Filtering Processing:
Wherein, x'=xcos θ+ysin θ and y'=-xsin θ+ycos θ;X and y represent pixel coordinate, λ tables
Show the wavelength of the sine curve factor, θ represents benchmark to the direction of parallel band,Phase offset is represented, σ represents Gauss
Factor standard is poor, γ representation space aspect rates.
The detection method of (note 12) according to note 11, wherein, filtered using different θ values travel directions
Ripple processes to strengthen the difference between each sub-block.
The detection method of (note 13) according to note 9 or 10, wherein, the statistical nature includes:Image
First moment and image second order square.
(note 14) detection method according to claim 13, wherein, carry out statistics using equation below special
The extraction levied:
Wherein, mean represents average, and var represents variance;M and N are respectively the real-time sub-block or the training
The height and width of sub-block, I () represents the pixel value of each pixel in the real-time sub-block or the training sub-block.
The detection method of (note 15) according to note 9, wherein, based on the sorter model being obtained ahead of time with
And the statistical nature carries out respectively type identification and includes to described each real-time sub-block:
For certain real-time sub-block, carried out based on the statistical nature of the sorter model being obtained ahead of time and the real-time sub-block
Score and obtain score value;
In the case where the score value is more than predetermined threshold value, the real-time sub-block is defined as into type of sports;Otherwise will
The real-time sub-block is defined as stationary kind.
The detection method of (note 16) according to note 9, wherein, the real-time figure is determined according to statistical result
As corresponding water state includes:
It is confirmed as the number of real-time sub-block of type of sports in the real time imaging more than in the case of predetermined threshold value,
The real time imaging is defined as into water sport state;Otherwise the real time imaging is defined as into water resting state.
(note 17) a kind of image processing equipment, including the detection of the water state as described in being attached 1 to 8 any one
Device.
Claims (10)
1. a kind of detection means of water state, it is characterised in that the detection means includes:
In real time partition unit, by the real time imaging of record water multiple real-time sub-blocks are divided into;
Real-Time Filtering unit, for each real-time sub-block distinguishes travel direction Filtering Processing, obtains that described each is real-time
The filter result of the different directions of sub-block;
Real-time characteristic extracting unit, according to the filter result of the different directions statistical nature is extracted respectively;
Type identification unit, based on the sorter model being obtained ahead of time and the statistical nature to described each real-time son
Block carries out respectively type identification;
Type statistics unit, counts to the type of the plurality of real-time sub-block of the real time imaging;And
Status determining unit, according to statistical result the corresponding water state of the real time imaging is determined.
2. detection means according to claim 1, wherein, the detection means also includes:
Training partition unit, by training image multiple training sub-blocks are divided into;
Training filter unit, for each training sub-block difference travel direction Filtering Processing, obtains described each training
The filter result of the different directions of sub-block;
Training characteristics extracting unit, according to the filter result of the different directions statistical nature is extracted respectively;And
Model signal generating unit, obtains described during the statistical nature is input to into support vector machine or artificial neural network
Sorter model.
3. detection means according to claim 1 and 2, wherein, the Real-Time Filtering unit or the training
Filter unit uses equation below travel direction Filtering Processing:
Wherein, x'=xcos θ+ysin θ and y'=-xsin θ+ycos θ;X and y represent pixel coordinate, λ tables
Show the wavelength of the sine curve factor, θ represents benchmark to the direction of parallel band,Phase offset is represented, σ represents Gauss
Factor standard is poor, γ representation space aspect rates.
4. detection means according to claim 3, wherein, the Real-Time Filtering unit or the training are filtered
Unit strengthens the difference between each sub-block using different θ value travel direction Filtering Processing;
The statistical nature includes:Image first moment and image second order square.
5. detection means according to claim 1 and 2, wherein, the real-time characteristic extracting unit or described
Training characteristics extracting unit carries out the extraction of statistical nature using equation below:
Wherein, mean represents average, and var represents variance;M and N are respectively the real-time sub-block or the training
The height and width of sub-block, I () represents the pixel value of each pixel in the real-time sub-block or the training sub-block.
6. detection means according to claim 1, wherein, the type identification unit includes:
Sub-block scoring unit, for certain real-time sub-block, based on the sorter model being obtained ahead of time and the real-time sub-block
Statistical nature scored and obtained score value;
Type determining units, in the case where the score value is more than predetermined threshold value, by the real-time sub-block fortune are defined as
Dynamic type;Otherwise the real-time sub-block is defined as into stationary kind.
7. detection means according to claim 1, wherein, the status determining unit specifically for:Institute
The number of the real-time sub-block for being confirmed as type of sports in real time imaging is stated more than in the case of predetermined threshold value, by the reality
When image be defined as water sport state;Otherwise the real time imaging is defined as into water resting state.
8. a kind of detection method of water state, it is characterised in that the detection method includes:
The real time imaging of record water is divided into into multiple real-time sub-blocks;
For each real-time sub-block distinguishes travel direction Filtering Processing, the different directions of each real-time sub-block are obtained
Filter result;
Statistical nature is extracted respectively according to the filter result of the different directions;
Class is carried out respectively to described each real-time sub-block based on the sorter model being obtained ahead of time and the statistical nature
Type is recognized;
The type of the plurality of real-time sub-block of the real time imaging is counted;And
The corresponding water state of the real time imaging is determined according to statistical result.
9. detection method according to claim 8, wherein, the real time imaging pair is determined according to statistical result
The water state answered includes:
It is confirmed as the number of real-time sub-block of type of sports in the real time imaging more than in the case of predetermined threshold value,
The real time imaging is defined as into water sport state;Otherwise the real time imaging is defined as into water resting state.
10. a kind of image processing equipment, it is characterised in that described image processing equipment includes such as claim 1 to 7
The detection means of the water state described in any one.
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510648508.7A CN106570854A (en) | 2015-10-09 | 2015-10-09 | Water state detection device and method and image processing device |
JP2016199042A JP2017073138A (en) | 2015-10-09 | 2016-10-07 | Water state detection device, method and image processing device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510648508.7A CN106570854A (en) | 2015-10-09 | 2015-10-09 | Water state detection device and method and image processing device |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106570854A true CN106570854A (en) | 2017-04-19 |
Family
ID=58507696
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201510648508.7A Pending CN106570854A (en) | 2015-10-09 | 2015-10-09 | Water state detection device and method and image processing device |
Country Status (2)
Country | Link |
---|---|
JP (1) | JP2017073138A (en) |
CN (1) | CN106570854A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110503807A (en) * | 2019-08-25 | 2019-11-26 | 武汉和时利自动化***工程有限公司 | Water body change monitoring system and method after a kind of flood floods |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102122354A (en) * | 2011-03-15 | 2011-07-13 | 上海交通大学 | Adaptive characteristic block selection-based gait identification method |
CN102564508A (en) * | 2011-12-14 | 2012-07-11 | 河海大学 | Method for implementing online tests of stream flow based on video images |
CN103116761A (en) * | 2013-03-07 | 2013-05-22 | 武汉大学苏州研究院 | Dynamic texture recognition method on basis of image sequence |
CN103413114A (en) * | 2013-05-17 | 2013-11-27 | 浙江大学 | Near-drowning behavior detection method based on support vector machine |
CN104165668A (en) * | 2014-08-18 | 2014-11-26 | 深圳市亿维锐创科技有限公司 | Water conservancy monitoring method for automatic pre-warning of water conservancy risks and system thereof |
-
2015
- 2015-10-09 CN CN201510648508.7A patent/CN106570854A/en active Pending
-
2016
- 2016-10-07 JP JP2016199042A patent/JP2017073138A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102122354A (en) * | 2011-03-15 | 2011-07-13 | 上海交通大学 | Adaptive characteristic block selection-based gait identification method |
CN102564508A (en) * | 2011-12-14 | 2012-07-11 | 河海大学 | Method for implementing online tests of stream flow based on video images |
CN103116761A (en) * | 2013-03-07 | 2013-05-22 | 武汉大学苏州研究院 | Dynamic texture recognition method on basis of image sequence |
CN103413114A (en) * | 2013-05-17 | 2013-11-27 | 浙江大学 | Near-drowning behavior detection method based on support vector machine |
CN104165668A (en) * | 2014-08-18 | 2014-11-26 | 深圳市亿维锐创科技有限公司 | Water conservancy monitoring method for automatic pre-warning of water conservancy risks and system thereof |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110503807A (en) * | 2019-08-25 | 2019-11-26 | 武汉和时利自动化***工程有限公司 | Water body change monitoring system and method after a kind of flood floods |
Also Published As
Publication number | Publication date |
---|---|
JP2017073138A (en) | 2017-04-13 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Huang et al. | Tracknet: A deep learning network for tracking high-speed and tiny objects in sports applications | |
CN109753885B (en) | Target detection method and device and pedestrian detection method and system | |
US20180107894A1 (en) | Image-based feature detection using edge vectors | |
CN103514432B (en) | Face feature extraction method, equipment and computer program product | |
CN106504271A (en) | Method and apparatus for eye tracking | |
US20180367764A1 (en) | Method and apparatus for sensing moving ball | |
CN109919008A (en) | Moving target detecting method, device, computer equipment and storage medium | |
CN107133627A (en) | Infrared light spot center point extracting method and device | |
CN107316322A (en) | Video tracing method and device and object identifying method and device | |
US11568682B2 (en) | Recognition of activity in a video image sequence using depth information | |
CN103218605B (en) | A kind of fast human-eye positioning method based on integral projection and rim detection | |
CN106296720A (en) | Human body based on binocular camera is towards recognition methods and system | |
CN102947863A (en) | Moving-object detection device | |
US10582179B2 (en) | Method and apparatus for processing binocular disparity image | |
CN107292318A (en) | Image significance object detection method based on center dark channel prior information | |
US11508144B2 (en) | Method and device for object detection | |
CN101655910A (en) | Training system, training method and detection method | |
CN105354856A (en) | Human matching and positioning method and system based on MSER and ORB | |
CN104881640A (en) | Method and device for acquiring vectors | |
CN104463855A (en) | Significant region detection method based on combination of frequency domain and spatial domain | |
CN114399480A (en) | Method and device for detecting severity of vegetable leaf disease | |
CN108596032B (en) | Detection method, device, equipment and medium for fighting behavior in video | |
CN104065954A (en) | Method for quickly detecting parallax scope of high-definition stereoscopic video | |
CN103413154A (en) | Human motion identification method based on normalized class Google measurement matrix | |
CN112037254A (en) | Target tracking method and related device |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
WD01 | Invention patent application deemed withdrawn after publication | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20170419 |