CN104123721A - Automatic fish school feeding control method based on video streaming image distributed dynamic characteristic technology - Google Patents
Automatic fish school feeding control method based on video streaming image distributed dynamic characteristic technology Download PDFInfo
- Publication number
- CN104123721A CN104123721A CN201410314811.9A CN201410314811A CN104123721A CN 104123721 A CN104123721 A CN 104123721A CN 201410314811 A CN201410314811 A CN 201410314811A CN 104123721 A CN104123721 A CN 104123721A
- Authority
- CN
- China
- Prior art keywords
- fish
- area
- image
- shoal
- feeding
- 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.)
- Granted
Links
Landscapes
- Image Analysis (AREA)
Abstract
The invention discloses an automatic fish school feeding control method based on a video streaming image distributed dynamic characteristic technology; compared with the prior art, the automatic fish school feeding control method solves the problem that in the existing automatic feeding control method does not reasonably utilize characteristic information of a fish school. The method comprises the following steps of: obtaining and processing an initial image, thresholding, analyzing frame difference, analyzing characteristics, computing an area change rate and a fish school eating speed, and controlling the state of a feeder according to the area change rate and the fish school eating speed. According to the automatic fish school feeding control method based on the video streaming image distributed dynamic characteristic technology, fish school eating parameters are identified and analyzed and a lower computer is controlled to feed by a real-time image processing method.
Description
Technical field
The present invention relates to fisheries management technical field of automatic control, is a kind of shoal of fish based on video streaming image distributed dynamic feature technology autocontrol method of throwing something and feeding specifically.
Background technology
It is popular problem in computer vision research field that distributed dynamic target signature detects always, detects and has a wide range of applications carrying out target identification, tracking and abnormal behaviour.For fields such as video monitoring, intelligent robot and intelligent automobile navigation, there is important effect.In addition, along with the fast development of agriculture technology of Internet of things, the distributed dynamic target detection increase in demand in aquatic products industry, animal husbandry, plays very large help to aspects such as intelligence production and abnormal behaviour monitorings.
In computer vision analysis, for distributed dynamic target signature, monitor this problem, traditional method have powerful connections null method, frame difference method, optical flow method etc.Although background null method computing method are simple and widespread use, it cannot meet in the obvious situation of needs, particularly illumination variation of the complicated environmental background changing, and easily causes larger error.Frame difference method is processed and is extracted dynamic object feature by two or three consecutive frame figure in video sequence being carried out to difference thresholding, and same algorithm calculates and is simple and easy to realize, and for variations such as environment, illumination, has very strong anti-noise ability.But frame difference method has certain restriction for distributed dynamic object variations frequency, easily occurs blind area, affects target detection accuracy.Optical flow method is utilized the time dependent light stream characteristic of dynamic object to set up optical flow constraint equation and is detected, then dynamic object complicated movement in actual conditions often, and the method calculation of complex, anti-noise ability is poor and hardware requirement is high, so applicability is little.
And cultivating for intensive fishing ground the operation of throwing something and feeding, shoal of fish feed parameter has certain regularity.Show as: after device for feeding fish bait start, the shoal of fish starts to assemble, feed shoal of fish area starts to increase, and reaches gradually maximum area value.Along with the carrying out of throwing something and feeding, after the feed fully of the part shoal of fish, start to take a stroll, at this moment shoal of fish area can dwindle gradually.From the angle of Optimal culture, when area is lower than the threshold value of setting and after continuing for some time, can close device for feeding fish bait and stop throwing something and feeding.Therefore, thus how solving distributed dynamic target signature in fishing ground environment detects the automatic control that realizes device for feeding fish bait and has become and be badly in need of the technical matters that solves.
Summary of the invention
To the object of the invention is rationally not utilize the problem of shoal of fish characteristic information in order solving in the existing autocontrol method of throwing something and feeding, to provide a kind of shoal of fish based on video streaming image distributed dynamic feature technology autocontrol method of throwing something and feeding to solve the problems referred to above.
To achieve these goals, technical scheme of the present invention is as follows:
The shoal of fish based on the video streaming image distributed dynamic feature technology autocontrol method of throwing something and feeding, comprises the following steps:
Initial pictures obtains and processes, and is determined the on time of device for feeding fish bait, and control slave computer and open device for feeding fish bait by host computer, and the image acquiring device Real-time Collection shoal of fish assembles feed image;
Thresholding is processed, and the sequence image after processing is carried out to thresholding processing, obtains target image W1, W2, W3;
Frame difference is analysed, and by target image W1, W2, the poor processing of W3 frame, obtains target image D1, D2;
Signature analysis, is used least square fitting target area, draws area parameters S1, S2;
Reference area rate of change and shoal of fish eating speed;
According to area change rate and shoal of fish eating speed, control device for feeding fish bait state.
Described initial pictures obtains and processes and comprises the following steps:
{ P1, P2, P3, P4 in sequence of video images ... Pn} obtains three frame video image sequence P1, P2, the P3 recurring, and the time interval between two two field pictures is Δ t;
After selected to the target area in image sequence P1, P2, P3, carry out gray scale conversion processing;
Carry out Gaussian smoothing denoising, its formula is as follows
Wherein: the standard deviation that σ is normal distribution;
Sequence image P1, P2, P3 after smoothing processing are made to histogram equalization and process, strengthen gradation of image value dynamic range, sequence image Q1, Q2 after being processed, Q3.
Described thresholding is treated to:
Sequence image Q1, Q2, Q3 are carried out to thresholding processing, obtain target image W1, W2, W3, its computing formula is as follows:
Wherein, thresh1, thresh2 are gray threshold amount, and W (x, y), Q (x, y) are positioned at the grey scale pixel value at (x, y) coordinate place for image.
Described frame difference is analysed and is comprised the following steps:
By target image W1, W2, W3, carry out frame difference and analyse, the formula that obtains target image D1, D2 is as follows;
Wherein, D1 (x, y), D2 (x, y) are respectively target image D1, D2 and are positioned at (x, y) pixel value of locating, W1 (x, y), W2 (x, y), W3 (x, y) be respectively image W1, W2, W3 is positioned at the pixel value that (x, y) locates, T is gray threshold amount;
D1, D2 are made to morphological dilations to be processed.
Described signature analysis step is as follows:
Behavioral characteristics point merger in target image D1, D2 is become to point set, get peripheral point set coordinate x
i, y
icarry out least square fitting processing, its formula is as follows:
Wherein, m is that peripheral point set comprises a sum, x
i, y
ibe the transverse and longitudinal coordinate of i point,
for objective function, δ
ifor minimizing parameter;
The ellipse formula going out according to least square fitting, calculates fitted ellipse length half shaft length a separately in D1, D2 figure
1, a
2and b
1, b
2, its formula is as follows:
Wherein,
a1, a
2and b
1, b
2the major semi-axis, the minor semi-axis size that are respectively fitted ellipse in D1, D2, x, y are coordinate parameters;
Calculate the value of area parameters S1, S2
S1=πa
1b
1,S2=πa
2b
2;
Wherein, S1, S2 are the ellipse area of D1, D2 Point Set matching gained ellipse.
Described reference area rate of change and shoal of fish eating speed comprise the following steps:
Reference area rate of change, its computing formula is as follows:
wherein,
for area change rate, Δ t is the time interval;
Calculate shoal of fish eating speed, its computing formula is as follows:
wherein,
for shoal of fish eating speed,
the area change rate that represents the three frame image sequence set in continuous two moment.
Described comprises the following steps according to area change rate and shoal of fish eating speed control device for feeding fish bait state:
When area parameter S 1, S2 positive change, slave computer is controlled device for feeding fish bait and is expanded bait throwing in area;
Work as area change rate
positive change and shoal of fish eating speed
during positive change, by slave computer, control device for feeding fish bait bait throwing in speed, its bait throwing in speed computing formula is as follows:
Wherein V is bait throwing in motor rotation speed, k
1for scale-up factor;
When area parameter S 1, S2 increase to maximal value, host computer records the shoal of fish and gathers Maximum Area S
max;
When area parameter S 1, S2 negative sense change and area change rate
when negative sense changes, slave computer is controlled device for feeding fish bait and is reduced bait throwing in speed and reduce bait throwing in area;
When area parameter S 1, S2 are less than the area threshold S of setting
min,
S
min=k
2s
max, k wherein
2for scale-up factor,
And the duration surpasses the time threshold T setting
delaytime, utilize slave computer to close device for feeding fish bait.
Beneficial effect
A kind of shoal of fish based on video streaming image distributed dynamic feature technology of the present invention autocontrol method of throwing something and feeding, the method for compared with prior art processing by realtime graphic is carried out discriminance analysis to shoal of fish feed parameter, controls slave computer bait throwing in.According to the area of dynamic range in shoal of fish feed image, the trend of area change rate, shoal of fish eating speed, the quantity of the judgement feed shoal of fish and feed behavioral parameters, thereby select targetedly bait throwing in speed, bait throwing in area and daily ration, feeding quantity to control, realize and precisely throw something and feed and feed back in time fish school behavior parameter.The distributed dynamic feature extraction the present invention is directed under the complex environment of fishing ground has obvious advantage, has effectively solved background null method error ratio larger, frame difference method poor accuracy, the weak deficiency that waits of optical flow method anti-noise ability.
Accompanying drawing explanation
Fig. 1 is method flow diagram of the present invention.
Fig. 2 is concrete steps figure of the present invention
Embodiment
For making that architectural feature of the present invention and the effect reached are had a better understanding and awareness, in order to preferred embodiment and accompanying drawing, coordinate detailed explanation, be described as follows:
As shown in Figure 1, a kind of shoal of fish based on video streaming image distributed dynamic feature technology of the present invention autocontrol method of throwing something and feeding, is characterized in that, comprises the following steps:
The first step, initial pictures obtains and processes, and is determined the on time of device for feeding fish bait, and control slave computer and open device for feeding fish bait by host computer, and the image acquiring device Real-time Collection shoal of fish assembles feed image.Host computer is opened device for feeding fish bait according on time control slave computer the shoal of fish is carried out to feeding, the shoal of fish starts to assemble feed, image acquiring device is video camera for example, records the shoal of fish and assembles feed image, and the shoal of fish assembles feed image and controls device for feeding fish bait state for post analysis shoal of fish feed situation in good time.Initial pictures obtains and processes specifically and comprises the following steps:
(1) according to input video stream picture, obtain sequence of video images.{ P1, P2, P3, P4 in sequence of video images ... Pn} obtains three frame video image sequence P1, P2, the P3 recurring, and the time interval between two two field pictures is Δ t.The present invention is based on continuous three frame image sequence set judges, as { P1, P2, P3}, { P2, P3, P4}, { P3, P4, P5} etc., contrast judgement (method of introducing in the 6th step) through a plurality of arrangement sets, thereby realize the automatic control of device for feeding fish bait, Δ t represents the mistiming between two two field pictures, is generally the inverse of camera frame per second.
(2) according to interested behavioral characteristics target area, the target area in selected target image sequence, i.e. shoal of fish feed aggregation zone.After selected to the target area in image sequence P1, P2, P3, carry out gray scale conversion processing, colour is changed into grey.
(3) in order to eliminate noise effect, more preserving edge effect, thus carrying out the processing of Gaussian Blur smoothing denoising, its formula is as follows
Wherein: the standard deviation that σ is normal distribution.Because in the time of calculating mean value, central point is at pixel initial point, so average is 0.The curved surface level line Shi Cong center that this formula generates starts to be the concentric circles of normal distribution.The value of each pixel is the weighted mean of adjacent pixel values around.The value of original pixels has maximum Gaussian distribution value, so have maximum weight, neighbor is along with far away apart from original pixels, and weight is less.
(4) sequence image P1, P2, P3 after smoothing processing are made to histogram equalization and process, sequence image Q1, Q2 after being processed, Q3.Because be subject to restriction and the interference of various conditions while processing image, the gray-scale value of image often with actual scenery Incomplete matching, this will directly affect the subsequent treatment of image.For far and near different mainly due to object, or camera when scanning sensitivity has compared with large buddhist meaning and produces gradation of image distortion, or because of the under-exposed situations such as grey scale change narrow range that make image, at this moment we adopt histogram equalization to process to strengthen the variation range of gray scale, contrast.
Second step, thresholding is processed, and the sequence image after processing is carried out to thresholding processing, obtains target image W1, W2, W3.Mainly utilize the gray difference between background and destination object in image sequence Q1, Q2, Q3.Under perfect condition, if the gray difference between background and destination object is very large, can utilize the gray feature of image to select one or more optimum gradation threshold values, and the gray-scale value of each pixel in image and threshold value are made comparisons, finally corresponding pixel is assigned in suitable class according to comparative result.The computing formula that its thresholding is processed is as follows:
Wherein, thresh1=25, thresh2=240 are gray threshold amount, and according to practical experience, gray threshold amount is 25,240 o'clock, and effect is better; W (x, y), Q (x, y) are positioned at the grey scale pixel value at (x, y) coordinate place for image.
The 3rd step, frame difference is analysed, and by target image W1, W2, the poor processing of W3 frame, obtains target image D1, D2.Gray-scale value by two two field picture corresponding pixel points before and after W1 and W2, W2 and W3 subtracts each other, and at ambient brightness, changes little in the situation that, if respective pixel degree differs very little, can think that target is static herein; If the grey scale change at place, target area is very large, can think that this is because target travel in image causes,
These pixels are identified out the position that just can obtain moving target.Its concrete steps are: by target image W1 and W2, W2 and W3, do the poor processing of frame, the formula that obtains target image D1, D2 is as follows;
Wherein, D1 (x, y), D2 (x, y) are respectively target image D1, D2 and are positioned at (x, y) pixel value of locating, W1 (x, y), W2 (x, y), W3 (x, y) be respectively image W1, W2, W3 is positioned at the pixel value that (x, y) locates, T is threshold value.
To target image D1, D2 makes morphological dilations and processes, and expansion process is content of the prior art, expands to refer to some images (or a part of region in image, be A) are carried out to convolution with core (being called B).Endorsing is arbitrary shape or size, and it has a definition reference point out separately.In most cases, core is that a little centre is with filled squares or the disk of reference point.Endorse to be considered as template or mask, expansion is the operation of asking local maximum.Core B and image convolution, calculate the pixel maximal value in the region that core B covers, and this maximal value assignment the pixel to reference point appointment.Will make like this highlight regions in image increase gradually.To merge in object with all background dots of object contact, target is increased, can replenish the cavity in target.Concrete operation step is: by each pixel in a structural element (being generally the size of 3*3) scan image, by each pixel in structural element and the pixel of its covering, make AND-operation, if be all 0, this pixel is 0, otherwise is 255.
The 4th step, signature analysis, is used least square fitting target area, draws area parameters S1, S2.Its signature analysis concrete steps are as follows:
(1) the behavioral characteristics point merger in target image D1, D2 is become to point set, target image D1, D2 are the results after comparing for target image W1, W2, W3, the variation that target image W1 and W2, W2 and W3 put position has more afterwards formed target image D1, D2, and its mid point is behavioral characteristics point.Get peripheral point set coordinate x
i, y
icarry out least square fitting processing, its formula is as follows:
Wherein, m is that peripheral point set comprises sum a little, x
i, y
ibe the transverse and longitudinal coordinate of i point,
for objective function, δ
ifor minimizing parameter.
(2) by dynamic object unique point point set, get peripheral two-dimentional point set and make least square method best-fit, can obtain Target ellipse.By Target ellipse, solve oval length semiaxis, thereby calculate fitted ellipse area, thereby draw dynamic object unique point point set region area parameter.If the direct indignant dynamic target signature region of drawing together out, polygonized structure, is difficult for calculating target area area often, to area parameters, calculates and increases difficulty.Therefore by the method for fitted ellipse, can solve easily dynamic object characteristic area area.
The ellipse formula going out according to least square fitting, calculates fitted ellipse length half shaft length a separately in D1, D2 figure
1, a
2and b
1, b
2, its formula is as follows:
Wherein, a
1, a
2and b
1, b
2the major semi-axis, the minor semi-axis size that are respectively fitted ellipse in D1, D2, x, y are coordinate parameters;
(3) calculate the value of area parameters S1, S2
S1=πa
1b
1,S2=πa
2b
2;
Wherein, S1, S2 are the ellipse area to D1, D2 Point Set matching gained ellipse.
The 5th step, reference area rate of change and shoal of fish eating speed.By area change rate, can judge the shoal of fish and carry out situation, the shoal of fish has just started feed and has belonged to state of aggregation, and feed belongs to disperse state to a certain extent afterwards, thereby can carry out based on area change rate the control judgement of device for feeding fish bait.Shoal of fish eating speed can be wanted for bait throwing in the adjustment of bait throwing in speed.Its reference area rate of change and shoal of fish eating speed specifically comprise the following steps:
(1) reference area rate of change, its computing formula is as follows:
wherein,
for area change rate, Δ t is the time interval;
(2) calculate shoal of fish eating speed, its computing formula is as follows:
wherein,
for shoal of fish eating speed,
the area change rate that represents the three frame image sequence set in continuous two moment.?
for the area change rate that P1, P2, P3} sequence image calculate,
for { the area change rate that P2, P3, P4} sequence image calculate.
The 6th step, controls device for feeding fish bait state according to area change rate and shoal of fish eating speed.Its concrete steps are as follows:
(1) when area parameter S 1, S2 positive change, illustrate in bait throwing in process, the shoal of fish assembles more and more, and formed gathering is also just increasing, so slave computer is controlled device for feeding fish bait expansion bait throwing in area.
(2) work as area change rate
positive change and shoal of fish eating speed
during positive change, by slave computer, control device for feeding fish bait bait throwing in speed, illustrate that the shoal of fish not only assembles more and more, and the eating speed after assembling is accelerated, need to increase the bait throwing in speed of device for feeding fish bait to satisfy the demand, can certainly increase daily ration, feeding quantity by controlling the valve of device for feeding fish bait.Bait throwing in speed computing formula is as follows:
Wherein V is bait throwing in motor rotation speed, k
1for scale-up factor.
(3), when area parameter S 1, S2 increase to maximal value, host computer records the shoal of fish and gathers Maximum Area S
max, when arriving S
maxafter, should occur that area change rate starts to be reduced near 0 value, shoal of fish eating speed is corresponding weakening also.Now, keep bait throwing in speed, bait throwing in area, daily ration, feeding quantity constant.
(4) when area parameter S 1, S2 negative sense change and area change rate
when negative sense changes, the feed of the declaratives shoal of fish fully starts to take a stroll, and area parameters starts to reduce, and area change rate is negative sense to be changed.Slave computer is controlled device for feeding fish bait and is reduced bait throwing in speed and reduce bait throwing in area;
(5) as area parameter S 1, S2, be less than the area threshold S of setting
min, S
min=K
2s
max, k wherein
2for scale-up factor, illustrate that shoal of fish feed finishes substantially, and the duration surpasses the time threshold T setting
delaytime, utilize slave computer to close device for feeding fish bait.
As further application of the present invention, can, by analysis long-time to fishing ground, big data quantity, show that shoal of fish feed Parameter Variation detects shoal of fish abnormal behaviour.Long-term throwing something and feeding in testing process, by quantitative contrast, can find whether the feed Maximum Area of fishing ground shoal of fish every day tends towards stability on year-on-year basis, it is normal etc. whether area change rate and eating speed change.When master system analyzes the obvious ANOMALOUS VARIATIONS of feed parameter, in time identification quantitative analysis, helps user to judge shoal of fish growth cycle, analyze fish school behavior abnormal parameters and feed back.
More than show and described ultimate principle of the present invention, principal character and advantage of the present invention.The technician of the industry should understand; the present invention is not restricted to the described embodiments; what in above-described embodiment and instructions, describe is principle of the present invention; the present invention also has various changes and modifications without departing from the spirit and scope of the present invention, and these changes and improvements all fall in claimed scope of the present invention.The protection domain that the present invention requires is defined by appending claims and equivalent thereof.
Claims (7)
1. the autocontrol method of throwing something and feeding of the shoal of fish based on video streaming image distributed dynamic feature technology, is characterized in that, comprises the following steps:
11) initial pictures obtains and processes, and is determined the on time of device for feeding fish bait, and control slave computer and open device for feeding fish bait by host computer, and the image acquiring device Real-time Collection shoal of fish assembles feed image;
12) thresholding is processed, and the sequence image after processing is carried out to thresholding processing, obtains target image W1, W2, W3;
13) frame difference is analysed, and by target image W1, W2, the poor processing of W3 frame, obtains target image D1, D2;
14) signature analysis, is used least square fitting target area, draws area parameters S1, S2;
15) reference area rate of change and shoal of fish eating speed;
16) according to area change rate and shoal of fish eating speed, control device for feeding fish bait state.
2. a kind of shoal of fish based on video streaming image distributed dynamic feature technology according to claim 1 autocontrol method of throwing something and feeding, is characterized in that, described initial pictures obtains and processes and comprises the following steps:
21) { P1, P2, P3, P4 in sequence of video images ... Pn} obtains three frame video image sequence P1, P2, the P3 recurring, and the time interval between two two field pictures is Δ t;
22) after selected to the target area in image sequence P1, P2, P3, carry out gray scale conversion processing;
23) carry out Gaussian smoothing denoising, its formula is as follows
Wherein: the standard deviation that σ is normal distribution;
24) sequence image P1, P2, P3 after smoothing processing are made to histogram equalization and process, strengthen gradation of image value dynamic range, sequence image Q1, Q2 after being processed, Q3.
3. a kind of shoal of fish based on video streaming image distributed dynamic feature technology according to claim 1 autocontrol method of throwing something and feeding, is characterized in that, described thresholding is treated to:
Sequence image Q1, Q2, Q3 are carried out to thresholding processing, obtain target image W1, W2, W3, its computing formula is as follows:
Wherein, thresh1, thresh2 are gray threshold amount, and W (x, y), Q (x, y) are positioned at the grey scale pixel value at (x, y) coordinate place for image.
4. a kind of shoal of fish based on video streaming image distributed dynamic feature technology according to claim 1 autocontrol method of throwing something and feeding, is characterized in that, described frame difference is analysed and comprised the following steps:
41) by target image W1, W2, W3, carry out frame difference and analyse, the formula that obtains target image D1, D2 is as follows;
Wherein, D1 (x, y), D2 (x, y) are respectively target image D1, D2 and are positioned at (x, y) pixel value of locating, W1 (x, y), W2 (x, y), W3 (x, y) be respectively image W1, W2, W3 is positioned at the pixel value that (x, y) locates, T is gray threshold amount;
42) D1, D2 being made to morphological dilations processes.
5. a kind of shoal of fish based on video streaming image distributed dynamic feature technology according to claim 1 autocontrol method of throwing something and feeding, is characterized in that, described signature analysis step is as follows:
51) the behavioral characteristics point merger in target image D1, D2 is become to point set, get peripheral point set coordinate xi, yi and carry out least square fitting processing, its formula is as follows:
Wherein, m is that peripheral point set comprises a sum, x
i, y
ibe the transverse and longitudinal coordinate of i point,
for objective function, δ
ifor minimizing parameter;
52) ellipse formula going out according to least square fitting, calculates fitted ellipse length half shaft length a separately in D1, D2 figure
1, a
2and b
1, b
2, its formula is as follows:
Wherein, a
1, a
2and b
1, b
2the major semi-axis, the minor semi-axis size that are respectively fitted ellipse in D1, D2, x, y are coordinate parameters;
53) calculate the value of area parameters S1, S2
S1=πa
1b
1,S2=πa
2B
2;
Wherein, S1, S2 are the ellipse area of D1, D2 Point Set matching gained ellipse.
6. a kind of shoal of fish based on video streaming image distributed dynamic feature technology according to claim 1 autocontrol method of throwing something and feeding, is characterized in that, described reference area rate of change and shoal of fish eating speed comprise the following steps:
61) reference area rate of change, its computing formula is as follows:
wherein,
for area change rate, Δ t is the time interval;
62) calculate shoal of fish eating speed, its computing formula is as follows:
wherein,
for shoal of fish eating speed,
the area change rate that represents the three frame image sequence set in continuous two moment.
7. a kind of shoal of fish based on video streaming image distributed dynamic feature technology according to claim 1 autocontrol method of throwing something and feeding, is characterized in that, described according to area change rate and shoal of fish eating speed, control device for feeding fish bait state and comprises the following steps:
71), when area parameter S 1, S2 positive change, slave computer is controlled device for feeding fish bait and is expanded bait throwing in area;
72) work as area change rate
positive change and shoal of fish eating speed
during positive change, by slave computer, control device for feeding fish bait bait throwing in speed, its bait throwing in speed computing formula is as follows:
Wherein V is bait throwing in motor rotation speed, k
1for scale-up factor;
73), when area parameter S 1, S2 increase to maximal value, host computer records the shoal of fish and gathers Maximum Area S
max;
74) when area parameter S 1, S2 negative sense change and area change rate
when negative sense changes, slave computer is controlled device for feeding fish bait and is reduced bait throwing in speed and reduce bait throwing in area;
75) as area parameter S 1, S2, be less than the area threshold S of setting
min,
S
min=k
2s
max, k wherein
2for scale-up factor,
And the duration surpasses the time threshold T setting
delaytime, utilize slave computer to close device for feeding fish bait.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410314811.9A CN104123721B (en) | 2014-07-02 | 2014-07-02 | A kind of shoal of fish based on video streaming image distributed dynamic feature technology feeds autocontrol method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410314811.9A CN104123721B (en) | 2014-07-02 | 2014-07-02 | A kind of shoal of fish based on video streaming image distributed dynamic feature technology feeds autocontrol method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN104123721A true CN104123721A (en) | 2014-10-29 |
CN104123721B CN104123721B (en) | 2017-08-01 |
Family
ID=51769118
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201410314811.9A Expired - Fee Related CN104123721B (en) | 2014-07-02 | 2014-07-02 | A kind of shoal of fish based on video streaming image distributed dynamic feature technology feeds autocontrol method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN104123721B (en) |
Cited By (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104542411A (en) * | 2014-12-19 | 2015-04-29 | 浙江大学 | Intelligent bait feeding device and method based on image processing technology |
CN106526112A (en) * | 2016-10-25 | 2017-03-22 | 浙江工业大学 | Water toxicity detection method based on fish activity analysis |
CN107047423A (en) * | 2017-04-28 | 2017-08-18 | 全椒县鮰鱼养殖专业合作社 | It is a kind of that feeding amount Channel-catfish fish culture systems are adjusted based on weight detecting |
CN107945223A (en) * | 2017-11-20 | 2018-04-20 | 成都霍比特科技有限公司 | A kind of rotating inclined automatic frog feed dispenser and video analysis method |
CN109360237A (en) * | 2018-12-07 | 2019-02-19 | 北京市水产科学研究所(国家淡水渔业工程技术研究中心) | A kind of prediction technique of total fish catches |
CN109640641A (en) * | 2016-09-05 | 2019-04-16 | 渔觅创新私人有限公司 | Feeding system and bait-throwing method |
CN111127411A (en) * | 2019-12-17 | 2020-05-08 | 北京深测科技有限公司 | Monitoring control method for fishery breeding |
CN111372060A (en) * | 2020-04-07 | 2020-07-03 | 北京海益同展信息科技有限公司 | Intelligent bait casting method and system and inspection vision device |
CN111783745A (en) * | 2020-08-06 | 2020-10-16 | 珠海南方利洋水产科技有限公司 | Fish health judgment method and device applied to pond culture and computer-readable storage medium |
CN112136741A (en) * | 2020-08-28 | 2020-12-29 | 盐城工学院 | Accurate feeding method for visual area |
CN114503946A (en) * | 2022-01-24 | 2022-05-17 | 海南大学 | Fishing ground feeding system and method based on high frame rate dynamic frame difference accurate identification |
CN115067257A (en) * | 2022-07-26 | 2022-09-20 | 金华市水产技术推广站(金华市水生动物疫病防控中心) | Bait casting method and system capable of accurately controlling feeding amount |
CN117581815A (en) * | 2023-12-28 | 2024-02-23 | 佛山市南海区杰大饲料有限公司 | Method and device for judging growth condition of industrial cultured fish |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102037354A (en) * | 2008-04-09 | 2011-04-27 | 科技研究局 | System and method for monitoring water quality |
EP2666354A1 (en) * | 2011-01-20 | 2013-11-27 | Nnt-systems Co., Ltd. | Automatic feeding system for underwater fish farm |
CN103809627A (en) * | 2013-12-02 | 2014-05-21 | 青岛蓝图文化传播有限公司市南分公司 | Aquaculture method |
-
2014
- 2014-07-02 CN CN201410314811.9A patent/CN104123721B/en not_active Expired - Fee Related
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102037354A (en) * | 2008-04-09 | 2011-04-27 | 科技研究局 | System and method for monitoring water quality |
EP2666354A1 (en) * | 2011-01-20 | 2013-11-27 | Nnt-systems Co., Ltd. | Automatic feeding system for underwater fish farm |
CN103809627A (en) * | 2013-12-02 | 2014-05-21 | 青岛蓝图文化传播有限公司市南分公司 | Aquaculture method |
Non-Patent Citations (4)
Title |
---|
C.NOBLE等: "The effect of differing self-feeding regimes on the growth, behaviour and fin damage of rainbow trout held in groups", 《AQUACULTURE》 * |
K.P.ANG等: "Control of feed dispensation in seacages using underwater video monitoring: effects on growth and food conversion", 《AQUACULTURAL ENGINEERING 16》 * |
王志勇等: "集中式自动投饵系统的研制", 《渔业现代化》 * |
范良衷等: "基于计算机视觉技术的运动鱼检测算法", 《农业工程学报》 * |
Cited By (16)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104542411A (en) * | 2014-12-19 | 2015-04-29 | 浙江大学 | Intelligent bait feeding device and method based on image processing technology |
CN109640641A (en) * | 2016-09-05 | 2019-04-16 | 渔觅创新私人有限公司 | Feeding system and bait-throwing method |
CN106526112A (en) * | 2016-10-25 | 2017-03-22 | 浙江工业大学 | Water toxicity detection method based on fish activity analysis |
CN107047423A (en) * | 2017-04-28 | 2017-08-18 | 全椒县鮰鱼养殖专业合作社 | It is a kind of that feeding amount Channel-catfish fish culture systems are adjusted based on weight detecting |
CN107945223A (en) * | 2017-11-20 | 2018-04-20 | 成都霍比特科技有限公司 | A kind of rotating inclined automatic frog feed dispenser and video analysis method |
CN109360237A (en) * | 2018-12-07 | 2019-02-19 | 北京市水产科学研究所(国家淡水渔业工程技术研究中心) | A kind of prediction technique of total fish catches |
CN109360237B (en) * | 2018-12-07 | 2019-06-14 | 北京市水产科学研究所(国家淡水渔业工程技术研究中心) | A kind of prediction technique of total fish catches |
CN111127411B (en) * | 2019-12-17 | 2023-08-01 | 北京深测科技有限公司 | Monitoring control method for fishery cultivation |
CN111127411A (en) * | 2019-12-17 | 2020-05-08 | 北京深测科技有限公司 | Monitoring control method for fishery breeding |
CN111372060A (en) * | 2020-04-07 | 2020-07-03 | 北京海益同展信息科技有限公司 | Intelligent bait casting method and system and inspection vision device |
CN111783745A (en) * | 2020-08-06 | 2020-10-16 | 珠海南方利洋水产科技有限公司 | Fish health judgment method and device applied to pond culture and computer-readable storage medium |
CN112136741A (en) * | 2020-08-28 | 2020-12-29 | 盐城工学院 | Accurate feeding method for visual area |
CN114503946A (en) * | 2022-01-24 | 2022-05-17 | 海南大学 | Fishing ground feeding system and method based on high frame rate dynamic frame difference accurate identification |
CN115067257A (en) * | 2022-07-26 | 2022-09-20 | 金华市水产技术推广站(金华市水生动物疫病防控中心) | Bait casting method and system capable of accurately controlling feeding amount |
CN117581815A (en) * | 2023-12-28 | 2024-02-23 | 佛山市南海区杰大饲料有限公司 | Method and device for judging growth condition of industrial cultured fish |
CN117581815B (en) * | 2023-12-28 | 2024-06-11 | 佛山市南海区杰大饲料有限公司 | Method and device for judging growth condition of industrial cultured fish |
Also Published As
Publication number | Publication date |
---|---|
CN104123721B (en) | 2017-08-01 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN104123721A (en) | Automatic fish school feeding control method based on video streaming image distributed dynamic characteristic technology | |
CN105354791B (en) | A kind of improved ADAPTIVE MIXED Gauss foreground detection method | |
Lameski et al. | Weed detection dataset with RGB images taken under variable light conditions | |
WO2015010451A1 (en) | Method for road detection from one image | |
DE102015207676A1 (en) | Method and device for obstacle detection based on a monocular camera | |
Bhadane et al. | Early pest identification in agricultural crops using image processing techniques | |
CN107358245B (en) | Method for detecting image collaborative salient region | |
CN104021533B (en) | A kind of real time imaging noise-reduction method and device | |
CN109242032B (en) | Target detection method based on deep learning | |
Zhang et al. | Individual leaf identification from horticultural crop images based on the leaf skeleton | |
CN105469423B (en) | A kind of online method for tracking target based on continuous attraction sub-neural network | |
CN106097320B (en) | Underwater sea cucumber image automatic segmentation method and device | |
Han et al. | Research and implementation of an improved canny edge detection algorithm | |
CN113349111A (en) | Dynamic feeding method, system and storage medium for aquaculture | |
CN117115210B (en) | Intelligent agricultural monitoring and adjusting method based on Internet of things | |
CN111882555B (en) | Deep learning-based netting detection method, device, equipment and storage medium | |
CN104598914A (en) | Skin color detecting method and device | |
CN105426928B (en) | A kind of pedestrian detection method based on Haar feature and EOH feature | |
CN115861721A (en) | Livestock and poultry breeding spraying equipment state identification method based on image data | |
CN109271864B (en) | Crowd density estimation method based on wavelet transformation and support vector machine | |
CN109034070B (en) | Blind separation method and device for replacement aliasing image | |
CN116524269A (en) | Visual recognition detection system | |
CN114943929A (en) | Real-time detection method for abnormal behaviors of fishes based on image fusion technology | |
CN104239903A (en) | QPSO (quantum-behaved particle swarm optimization) algorithm based image edge detection method | |
CN114155590A (en) | Face recognition method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
CF01 | Termination of patent right due to non-payment of annual fee | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20170801 Termination date: 20190702 |