CN110796632B - Pig counting device - Google Patents
Pig counting device Download PDFInfo
- Publication number
- CN110796632B CN110796632B CN201910693494.9A CN201910693494A CN110796632B CN 110796632 B CN110796632 B CN 110796632B CN 201910693494 A CN201910693494 A CN 201910693494A CN 110796632 B CN110796632 B CN 110796632B
- Authority
- CN
- China
- Prior art keywords
- area
- pig
- image
- crowded
- infrared image
- 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.)
- Active
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
-
- 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
- G06T2207/10—Image acquisition modality
- G06T2207/10048—Infrared image
-
- 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
- G06T2207/20084—Artificial neural networks [ANN]
-
- 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
- G06T2207/20212—Image combination
- G06T2207/20221—Image fusion; Image merging
-
- 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
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30242—Counting objects in image
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
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)
Abstract
An embodiment of the present invention relates to a pig counting device, comprising: an image acquisition unit for acquiring an infrared image and a visible light image of the same area photographed at the same time; an infrared image crowded area determining unit for determining a crowded area of pigs in an infrared image, wherein the crowded area of pigs refers to an area where a plurality of pigs are crowded with each other; a visible light image crowded area determining unit for determining a crowded area of pigs in a visible light image; a common congestion area determination unit that determines a common congestion area in the visible light image and the infrared image; the fusion unit fuses the visible light image part and the infrared image part corresponding to the common crowded area part to obtain a fused image; the identification unit is used for carrying out convolutional neural network identification on the fused images and determining the number of pigs in the common crowded area; and a total counting unit for counting the swinery according to the identification result of the identification unit.
Description
Technical Field
The invention relates to a pig counting device.
Background
Pig counting has a lot of practical significance. The earmark of the pig can be used for identifying and counting pigs, but the earmark is easy to damage the pigs and is easy to drop or change. On the other hand, if the number of pigs is small, the image recognition method can be adopted to recognize the faces of the pigs in the image, so that the number of the pigs can be recognized.
However, the number of pigs in the pig farm is increasing at present, and local crowding occurs, and in this case, the accuracy of current image recognition needs to be improved.
Disclosure of Invention
The present invention has been made in view of the above problems occurring in the prior art, and provides a pig counting method and apparatus which alleviates or overcomes the above disadvantages of the prior art, providing at least one advantageous option.
According to one aspect of the present invention, there is provided a pig counting device 1, comprising: an image acquisition unit for acquiring an infrared image and a visible light image of the same area photographed at the same time; an infrared image crowded area determining unit for determining a crowded area of pigs in an infrared image, wherein the crowded area of pigs refers to an area where a plurality of pigs are crowded with each other; a visible light image crowded area determining unit for determining a crowded area of pigs in a visible light image; a common congestion area determination unit that determines a common congestion area in the visible light image and the infrared image, the common congestion area being a pig congestion area identified as a pig congestion area in both the visible light image and the infrared image; the fusion unit is used for fusing the visible light image part and the infrared image part corresponding to the common crowded area part to obtain a fused image; the identification unit is used for carrying out convolutional neural network identification on the fused images and determining the number of pigs in the common crowded area; and a total counting unit for counting the swinery according to the identification result of the identification unit.
According to one embodiment, the fusion unit performs NSCT transform on the visible light image portion and the infrared image portion to be fused to obtain corresponding low-frequency subband coefficients and high-frequency subband coefficients, performs low-frequency subband coefficient fusion and high-frequency subband coefficient fusion according to a fusion rule, performs NSCT inverse transform to obtain a fused image,
wherein the low frequency subband coefficient fusion is performed as follows:
first, a characteristic Af common to low-frequency subbands of an infrared image portion and a visible image portion is obtained comm ：
Af comm ＝min(Af inf ,Af nature )
Wherein Af is inf Features representing low frequency subbands of the infrared image portion, af nature Features representing low frequency subbands of the visible image portion;
then, characteristic features of the infrared image section are obtained:
Af spcial ＝Af inf -Af comm
next, a low frequency subband of the fused image is generated
σ(Af inf ) Representing the regional variance, σ (Af, of the infrared image nature ) Representing the regional variance of the natural image.
According to one embodiment, the identification unit comprises a common crowded area identification unit and a visible light image pig crowded area identification unit; the common crowding area identification unit adopts Ic-CNN, SANet or CSRNet to carry out common crowding area identification, and identifies the number of designated pig outlines, wherein the designated pig outlines are outlines determined according to all or part of the forelimbs, the hindlimbs and the forelimbs of the pigs; the visible light image pig crowded area identifying unit identifies and counts other pig crowded areas except the common crowded area in the visible light image, firstly, an infrared image corresponding area corresponding to each pig crowded area in the infrared image is determined, and pigs in each infrared image corresponding area are identified and counted as the number of pigs in each pig crowded area in the visible light image.
According to one embodiment, the total counting unit counts the areas outside the crowded areas of the pigs in the visible light image based on a pig face recognition algorithm, and obtains the total counting result by utilizing the recognition result of the recognition unit of the crowded areas of the pigs in the visible light image and the recognition result of the common crowded area recognition unit.
According to one embodiment, the visible light image crowded area determining unit determines a crowded area of pigs in the visible light image according to the accuracy of the pig face recognition algorithm.
According to one embodiment, the fusion unit performs NSCT transform on the visible light image portion and the infrared image portion to be fused to obtain a corresponding low-frequency subband coefficient and high-frequency subband coefficient, then performs low-frequency subband coefficient fusion and high-frequency subband coefficient fusion according to a fusion rule, and then performs NSCT inverse transform to obtain a fused image, wherein the low-frequency subband coefficient fusion is performed as follows:
first, a characteristic Af common to low-frequency subbands of an infrared image portion and a visible image portion is obtained comm ：
Af comm ＝min(Af inf ,Af nature )
Wherein Af is inf Features representing low frequency subbands of the infrared image portion, af nature Features representing low frequency subbands of the visible image portion;
then, characteristic features of the infrared image section are obtained:
Af spcial ＝Af inf -Af comm
next, low frequency subbands of the fused image are generated:
σ(Af inf ) Representing the regional variance, σ (Af, of the infrared image nature ) And expressing the regional variance of the natural image, wherein beta is the ratio of the confidence coefficient of the pig face recognition algorithm to the confidence coefficient of the convolutional neural network recognition algorithm adopted by the common crowded region recognition unit.
According to one embodiment, the identification unit comprises a common crowded area identification unit and an infrared image pig crowded area identification unit; the common crowding area identification unit adopts Ic-CNN, SANet or CSRNet to carry out common crowding area identification, and identifies the number of designated pig outlines, wherein the designated pig outlines are outlines determined according to all or part of the forelimbs, the hindlimbs and the forelimbs of the pigs; the infrared image pig crowded area identification unit identifies and counts other pig crowded areas except the common crowded area in the infrared image, firstly, a visible light image corresponding area corresponding to each pig crowded area in the visible light image is determined, and pigs in each visible light image corresponding area are identified and counted as the number of pigs in each corresponding pig crowded area in the corresponding infrared image.
According to one embodiment, the total counting unit counts the areas outside the crowded area of the pigs in the infrared image based on a pig contour recognition algorithm, and obtains the total recognition result by using the recognition result of the infrared image crowded area recognition unit and the recognition result of the common crowded area recognition unit.
According to one embodiment, the infrared image crowded area determining unit determines a crowded area of pigs in the visible light image according to the accuracy of the pig profile recognition algorithm.
According to one embodiment, the fusion unit performs NSCT transform on the visible light image portion and the infrared image portion to be fused to obtain corresponding low-frequency subband coefficients and high-frequency subband coefficients, performs low-frequency subband coefficient fusion and high-frequency subband coefficient fusion according to a fusion rule, performs NSCT inverse transform to obtain a fused image,
wherein the low frequency subband coefficient fusion is performed as follows:
first, a characteristic Af common to low-frequency subbands of an infrared image portion and a visible image portion is obtained comm ：
Af comm ＝min(Af inf ,Af nature )
Wherein Af is inf Features representing low frequency subbands of the infrared image portion, af nature Features representing low frequency subbands of the visible image portion;
then, characteristic features of the infrared image section are obtained:
Af spcial ＝Af inf -Af comm
next, low frequency subbands of the fused image are generated:
σ(Af inf ) Representing the regional variance, σ (Af, of the infrared image nature ) And expressing the regional variance of the natural image, wherein beta is the ratio of the confidence coefficient of the pig contour recognition algorithm to the confidence coefficient of the convolutional neural network recognition algorithm adopted by the common crowded region recognition unit.
Drawings
The drawings are only schematic and are non-limiting to the scope of the invention.
Fig. 1 shows a block diagram of a pig counting device according to an embodiment of the invention.
Fig. 2 shows a schematic block diagram of an identification unit according to an embodiment.
Detailed Description
The following detailed description of embodiments of the invention, taken in conjunction with the accompanying drawings, is given by way of illustration and not limitation of the scope of the invention.
Fig. 1 shows a block diagram of a pig counting device according to an embodiment of the invention. As shown in fig. 1, according to an embodiment of the present invention, a pig counting device includes: an image acquisition unit 10 for acquiring an infrared image and a visible light image of the same area photographed at the same time, which can be input or received from the outside, and which can be acquired by an infrared camera and a visible light camera; an infrared image crowded area determining unit 20 that determines a crowded area of pigs in an infrared image, the crowded area of pigs being an area where a plurality of pigs are crowded with each other, for example, a pig slot area when a person is on robbed; a visible light image crowded area determination unit 30 that determines a crowded area of pigs in a visible light image; a common congestion area determination unit 40 that determines a common congestion area in the visible light image and the infrared image, the common congestion area being a pig congestion area identified as a pig congestion area in both the visible light image and the infrared image; a fusion unit 50 for fusing the visible light image and the infrared image of the common crowded area part to obtain a fused image; the recognition unit 60 performs Convolutional Neural Network (CNN) recognition on the fused images, and determines the number of pigs in the common crowded area; and a total counting unit 70 for counting the pig group according to the identification result of the identification unit 60.
The infrared image congestion area determining unit 20 may determine the congestion area of pigs according to the accuracy of the infrared image recognition algorithm, that is, only the range that will cause the reduction of the recognition accuracy is listed in the congestion area of pigs, so that the minimum number of pigs in the congestion area may be different according to the algorithm.
The visible light image crowded area determining unit 30 may determine according to the accuracy of the visible light image recognition algorithm when determining the crowded area of pigs, i.e., only the range that will cause the degradation of the recognition accuracy is listed in the crowded area of pigs, and thus the minimum number of pigs in the crowded area may be different from algorithm to algorithm and may be different from the minimum number of pigs in the crowded area of pigs in the infrared image.
The fusing unit 50 may perform the fusing using various methods known at present or known in the future. According to one embodiment, the visible light image and the infrared image may be first transformed NSCT (Nonsubsampled Contourlet Transform) to obtain the corresponding low-frequency subband coefficient and high-frequency subband coefficient, and then coefficient fused according to the fusion rule, and then inverse NSCT transformation is performed to obtain the fused image. The infrared image of the pig is characterized in that on one hand, the background difference between the pig body and the pig house is obvious, and meanwhile, the temperature difference exists between each part of the pig body, mainly the difference between the ear part and part of the mouth of the pig and the pig body or more. The gray level distribution of each part of the pig body in the visible light image is close to that of the pig body, but the difference between the pig body and the background is smaller. In the conventional fusion, feature extraction is generally performed to perform face recognition and posture recognition of pigs. However, the inventors found that the enhanced texture and detail feature points were not particularly significant for the total identification of pigs, but rather were slow to operate. An algorithm is thus provided that makes use of, inter alia, gray-level uniform portions of the infrared image to better distinguish the corresponding portions of the visible image from the background. The high frequency coefficients mainly represent texture information and details, and can be fused by various methods in the field. The low frequency subband coefficient fusion may be performed as follows:
first, an infrared image is obtained andfeatures Af common to low frequency subbands of visible light images comm ：
Af comm ＝min(Af inf ,Af nature )
Wherein Af is inf Features representing low frequency subbands of an infrared image, af nature Features representing low frequency subbands of the visible light image.
Then, characteristic features of the infrared image are obtained:
Af spcial ＝Af inf -Af comm
next, a low frequency subband of the fused image is generated
σ(Af inf ) Representing the regional variance, σ (Af, of the infrared image nature ) Representing the regional variance of the natural image.
By adopting the method, the later identification speed is higher, and the pig count is better.
Fig. 2 shows a schematic diagram of an identification unit according to an embodiment. As shown in fig. 2, the identifying unit 60 includes a common crowded area identifying unit 601, a visible light image pig crowded area identifying unit 602, and an infrared image pig crowded area identifying unit 603 according to one embodiment.
The common congestion area identification unit 601 may use Ic-CNN, SANet, CSRNet and the like to realize the common congestion area density map and swinery identification. When the crowd is identified, the Ic-CNN is thinned from a low-resolution density map to a high-resolution density map, and the rest methods are used for extracting multi-scale head characteristic information. The invention is special for identifying pig herds. Ic-CNN, SANet, and CSRNet are examples of convolutional neural network identification of the present invention, performing end-to-end identification, identifying the number of designated pig contours, which are contours determined from all or part of the pig between the forelimbs and hindlimb forelimbs.
According to one embodiment, a single-channel picture of the same size as the original image is first generated, wherein all pixels take 0 s, then the point with the designated pig outline is marked as 1, and then the picture is processed through Gaussian filtering, and the formed picture is a density picture. It should be noted that unlike the identification of crowd density maps, the pig head portion of the pig tends to be relatively warm, and through the above fusion, some portions of the pig head portion (e.g., the ears) are obscured, where the designated pig contour may be selected in whole or in part between the forelimbs and hindlimb forelimbs. In one embodiment, a portion of the first distance from behind the anterior limb to a second distance in front of the posterior limb may be selected. The first distance and the second distance can be determined according to the breeds, growing periods and seasons of the pigs, so that the part which is most distinguished from the pigsty background is selected. The pig profile at the time of identification is determined based on the portion.
The visible light image pig crowded area identifying unit 602 identifies and counts other pig crowded areas except for the common crowded area in the visible light image, and according to one embodiment, it first determines an infrared image corresponding area in the infrared image corresponding to each pig crowded area in the visible light image, identifies and counts pigs in each infrared image corresponding area, as the number of pigs in each pig crowded area in the visible light image.
Similarly, the infrared image pig crowded area identifying unit 603 identifies and counts other pig crowded areas except for the common crowded area in the infrared image, and according to one embodiment, it first determines a visible light image corresponding area corresponding to each of the pig crowded areas in the visible light image, identifies and counts pigs in each of the visible light image corresponding areas, as the number of pigs in each of the corresponding pig crowded areas in the corresponding infrared image.
According to one embodiment, the total counting unit 70 may identify an area other than the crowded area of the swine in the visible light image, and obtain the total counting result using the identification result of the visible light image swine crowded area identification unit 602 and the identification result of the common crowded area identification unit 601. According to one embodiment, the total counting unit 70 may perform counting by using a method based on the face recognition of pigs when recognizing areas other than the crowded area of pigs in the visible light image. Based on the identified faces of pigs and counting them further, the accuracy can be improved without any problem.
In this case, when generating the low-frequency subband of the fused image, at Af nature The coefficient beta is added before, and the beta is the ratio of the confidence coefficient of the pig face recognition algorithm to the confidence coefficient of the convolutional neural network recognition algorithm adopted by the common congestion area recognition unit 601. I.e.
According to this method, a degree of post feedback is achieved due to the confidence of the latter algorithm being considered in fusion, thus allowing for greater accuracy in the identification of commonly congested areas.
According to another embodiment, the total counting unit 70 may identify the area outside the crowded area of the pigs in the infrared image, and obtain the total identification result (i.e. the sum) by using the identification result of the infrared image crowded area identification unit 603 and the identification result of the common crowded area identification unit 601. The total counting unit 70 may use a counting based on a pig contour recognition algorithm when recognizing areas other than the crowded area of the pig in the infrared image.
In this case, when generating the low-frequency subband of the fused image, at Af nature The coefficient beta is added before, and the beta is the ratio of the confidence coefficient of the pig profile recognition algorithm to the confidence coefficient of the convolutional neural network recognition algorithm adopted by the common congestion area recognition unit 601. I.e.
Thus, as can be seen from the above, the infrared image pig crowded area identification unit 603 and the visible image pig crowded area identification unit 602 need not be provided at the same time.
According to one embodiment, the total counting unit 70 may identify the area outside the crowded area of the swine in the infrared image, and obtain the first result by using the identification result of the infrared image crowded area identification unit 603 and the identification result of the common crowded area identification unit 601, and at the same time, the total counting unit 70 may identify the area outside the crowded area of the swine in the visible image, obtain the second counting result by using the identification result of the visible image crowded area identification unit 602 and the identification result of the common crowded area identification unit 601, and average or weighted average the first result and the second result to obtain the total counting result.
According to the method provided by the invention, the common crowded area which is most difficult to identify is firstly cut, and the CNN end-to-end neural network identification method for identifying the common crowded area is used for reference, so that the overall task amount is reduced, and the overall identification speed is improved, on the other hand, as the identification is only carried out on the common crowded area, even if the identification precision is lower than that of the uncongested part of the visible light image or the infrared image, the influence on the overall identification precision is relatively small, and the overall accuracy improvement can be ensured. Further, if the region where pigs are crowded only in the infrared image or the region where pigs are crowded only in the visible light image is identified by utilizing the characteristic that the region is not crowded in the other image, the accuracy and efficiency of identification can be improved as well. The identification of uncongested portions of the visible and infrared images may be selected from the best accuracy identification algorithms now known or later known.
The pig counting device may be implemented by a computer (or in combination with a camera, etc.) comprising storage means and computing means (CPU, etc.). The memory of the computer stores computer software which, when executed (including the case of compiled execution), can cause the computer to implement the face recognition device of the present invention.
One aspect of the invention also includes the computer software and a medium storing the computer software.
It should be noted that the described embodiments are only some embodiments of the invention, and not all embodiments. Any and all other embodiments, which are within the scope of the claims of the present invention, are based on the inventive concept.
Claims (6)
1. A pig counting device, comprising:
an image acquisition unit for acquiring an infrared image and a visible light image of the same area photographed at the same time;
an infrared image crowded area determining unit for determining a crowded area of pigs in an infrared image, wherein the crowded area of pigs refers to an area where a plurality of pigs are crowded with each other and is an area which can cause the reduction of the recognition accuracy of an infrared image pig contour recognition algorithm;
a visible light image crowded area determining unit for determining a crowded area of pigs in the visible light image, wherein the crowded area of pigs refers to an area where a plurality of pigs are crowded with each other and is an area which can cause the reduction of the recognition accuracy of a visible light image pig face recognition algorithm;
a common congestion area determination unit that determines a common congestion area of the visible light image and the infrared image, the common congestion area being a pig congestion area identified as a pig congestion area in both the visible light image and the infrared image;
the fusion unit is used for fusing the visible light image part and the infrared image part corresponding to the common crowded area part to obtain a fused image;
the identification unit is used for carrying out convolutional neural network identification on the fusion image and determining the number of pigs in a common crowded area; and
a total counting unit for counting the swinery according to the identification result of the identification unit,
wherein the identification unit comprises a common crowded area identification unit and a visible light image pig crowded area identification unit,
the visible light image pig crowded area identifying unit identifies and counts other pig crowded areas except the common crowded area in the visible light image, firstly, an infrared image corresponding area corresponding to each pig crowded area in the infrared image is determined, then, pigs in each infrared image corresponding area are identified and counted, and the number of pigs in each pig crowded area in the visible light image is taken as the number of pigs in each pig crowded area in the visible light image,
wherein the common congestion area identification unit identifies the common congestion area by Ic-CNN, SANet or CSRNet, identifies the number of designated pig contours which are contours determined according to all or part of the forelimbs, hind limbs and forelimbs of the pigs,
and the total counting unit counts the areas except for the crowded areas of the pigs in the visible light image based on a pig face recognition algorithm, and obtains the total counting result by utilizing the recognition result of the crowded area recognition unit of the pigs in the visible light image and the recognition result of the common crowded area recognition unit.
2. The pig counting device according to claim 1, wherein the fusion unit performs NSCT transform on the visible light image portion and the infrared image portion to be fused to obtain corresponding low-frequency subband coefficients and high-frequency subband coefficients, performs low-frequency subband coefficient fusion and high-frequency subband coefficient fusion according to a fusion rule, performs NSCT inverse transform to obtain a fused image,
wherein the low frequency subband coefficient fusion is performed as follows:
first, a characteristic Af common to low-frequency subbands of an infrared image portion and a visible image portion is obtained comm ：
Af comm ＝min(Af inf ,Af nature )
Wherein Af is inf Features representing low frequency subbands of the infrared image portion, af nature Features representing low frequency subbands of the visible image portion;
then, the characteristic features of the infrared image section are obtained:
Af spcial ＝Af inf -Af comm
next, a low frequency subband Af of the fused image is generated fuse，
σ(Af inf ) Representing the regional variance, σ (Af, of the infrared image nature ) Representing the regional variance of the natural image.
3. The pig counting device according to claim 1, wherein,
the fusion unit carries out NSCT transformation on the visible light image part and the infrared image part to be fused to obtain corresponding low-frequency subband coefficients and high-frequency subband coefficients, then carries out low-frequency subband coefficient fusion and high-frequency subband coefficient fusion according to a fusion rule, then carries out NSCT inverse transformation to obtain a fused image,
wherein the low frequency subband coefficient fusion is performed as follows:
first, a characteristic Af common to low-frequency subbands of an infrared image portion and a visible image portion is obtained comm ：
Af comm ＝min(Af inf ,Af nature )
Wherein Af is inf Features representing low frequency subbands of the infrared image portion, af nature Features representing low frequency subbands of the visible image portion;
then, characteristic features of the infrared image section are obtained:
Af spcial ＝Af inf -Af comm
next, low frequency subbands of the fused image are generated:
σ(Af inf ) Representing the regional variance, σ (Af, of the infrared image nature ) Representing the regional variance of natural images, wherein beta is the confidence of the pig face recognition algorithm and the common crowded region recognition unit adoptsThe ratio of confidence levels of the convolutional neural network recognition algorithm used.
4. The pig counting device according to claim 1, wherein,
the infrared image pig crowded area identification unit identifies and counts other pig crowded areas except the common crowded area in the infrared image, firstly, a visible light image corresponding area corresponding to each pig crowded area in the visible light image is determined, then, pigs in each visible light image corresponding area are identified and counted, and the number of pigs in each corresponding pig crowded area in the corresponding infrared image is used as the number of pigs in each corresponding pig crowded area.
5. The pig counting device according to claim 4, wherein the total counting unit counts the areas outside the crowded areas of the pigs in the infrared image based on a pig contour recognition algorithm, and obtains the total recognition result by using the recognition result of the infrared image pig crowded area recognition unit and the recognition result of the common crowded area recognition unit.
6. The pig counting device according to claim 5, wherein the fusion unit performs NSCT transform on the visible light image portion and the infrared image portion to be fused to obtain corresponding low-frequency subband coefficients and high-frequency subband coefficients, performs low-frequency subband coefficient fusion and high-frequency subband coefficient fusion according to a fusion rule, performs NSCT inverse transform to obtain a fused image,
wherein the low frequency subband coefficient fusion is performed as follows:
first, a characteristic Af common to low-frequency subbands of an infrared image portion and a visible image portion is obtained comm ：
Af comm ＝min(Af inf ,Af nature )
Wherein Af is inf Features representing low frequency subbands of the infrared image portion, af nature Features representing low frequency subbands of the visible image portion;
then, characteristic features of the infrared image section are obtained:
Af spcial ＝Af inf -Af comm
next, low frequency subbands of the fused image are generated:
σ(Af inf ) Representing the regional variance, σ (Af, of the infrared image nature ) And expressing the regional variance of the natural image, wherein beta is the ratio of the confidence coefficient of the pig contour recognition algorithm to the confidence coefficient of the convolutional neural network recognition algorithm adopted by the common crowded region recognition unit.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910693494.9A CN110796632B (en) | 2019-07-30 | 2019-07-30 | Pig counting device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910693494.9A CN110796632B (en) | 2019-07-30 | 2019-07-30 | Pig counting device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110796632A CN110796632A (en) | 2020-02-14 |
| CN110796632B true CN110796632B (en) | 2023-08-11 |
Family
ID=69426937
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910693494.9A Active CN110796632B (en) | 2019-07-30 | 2019-07-30 | Pig counting device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110796632B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112200003B (en) * | 2020-09-14 | 2024-02-20 | 浙江大华技术股份有限公司 | Method and device for determining feed feeding amount in pig farm |
| CN112070791A (en) * | 2020-09-21 | 2020-12-11 | 深圳喜为智慧科技有限公司 | Method and system for improving accuracy and efficiency of animal husbandry individual points |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP4652480B1 (en) * | 2010-07-02 | 2011-03-16 | オオクマ電子株式会社 | Device for counting the number of drugs in a medicine package |
| CN102637297A (en) * | 2012-03-21 | 2012-08-15 | 武汉大学 | Visible light and infrared image fusion method based on Curvelet transformation |
| CN103076073A (en) * | 2011-10-26 | 2013-05-01 | 因诺威泰克有限公司 | Device for determining weight and number of chick and sorting device comprising the same |
| CN105719263A (en) * | 2016-01-22 | 2016-06-29 | 昆明理工大学 | Visible light and infrared image fusion algorithm based on NSCT domain bottom layer visual features |
| CN106780419A (en) * | 2016-11-28 | 2017-05-31 | 深圳汇通智能化科技有限公司 | A kind of missing child automatic recognition system based on big data |
| CN107230196A (en) * | 2017-04-17 | 2017-10-03 | 江南大学 | Infrared and visible light image fusion method based on non-down sampling contourlet and target confidence level |
| TWI611374B (en) * | 2017-05-04 | 2018-01-11 | Chunghwa Telecom Co Ltd | Gender and age identification method for vertical image flow counting |
| CN107977924A (en) * | 2016-10-21 | 2018-05-01 | 杭州海康威视数字技术股份有限公司 | A kind of image processing method based on dual sensor imaging, system |
| CN108171162A (en) * | 2017-12-27 | 2018-06-15 | 重庆交通开投科技发展有限公司 | Crowded degree detection method, apparatus and system |
| CN108207653A (en) * | 2018-01-31 | 2018-06-29 | 佛山市神风航空科技有限公司 | A kind of grazing system based on unmanned plane |
| CN109461151A (en) * | 2018-11-05 | 2019-03-12 | 上海睿畜电子科技有限公司 | A kind of method, apparatus and system that livestock number is checked |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2010127023A1 (en) * | 2009-05-01 | 2010-11-04 | Spicola Tool, Llc | Remote contactless stereoscopic mass estimation system |
| US8720382B2 (en) * | 2010-08-31 | 2014-05-13 | Technologies Holdings Corp. | Vision system for facilitating the automated application of disinfectant to the teats of dairy livestock |
| WO2017120384A1 (en) * | 2016-01-08 | 2017-07-13 | Flir Systems, Inc. | Thermal-image based object detection and heat map generation systems and methods |
-
2019
- 2019-07-30 CN CN201910693494.9A patent/CN110796632B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP4652480B1 (en) * | 2010-07-02 | 2011-03-16 | オオクマ電子株式会社 | Device for counting the number of drugs in a medicine package |
| CN103076073A (en) * | 2011-10-26 | 2013-05-01 | 因诺威泰克有限公司 | Device for determining weight and number of chick and sorting device comprising the same |
| CN102637297A (en) * | 2012-03-21 | 2012-08-15 | 武汉大学 | Visible light and infrared image fusion method based on Curvelet transformation |
| CN105719263A (en) * | 2016-01-22 | 2016-06-29 | 昆明理工大学 | Visible light and infrared image fusion algorithm based on NSCT domain bottom layer visual features |
| CN107977924A (en) * | 2016-10-21 | 2018-05-01 | 杭州海康威视数字技术股份有限公司 | A kind of image processing method based on dual sensor imaging, system |
| CN106780419A (en) * | 2016-11-28 | 2017-05-31 | 深圳汇通智能化科技有限公司 | A kind of missing child automatic recognition system based on big data |
| CN107230196A (en) * | 2017-04-17 | 2017-10-03 | 江南大学 | Infrared and visible light image fusion method based on non-down sampling contourlet and target confidence level |
| TWI611374B (en) * | 2017-05-04 | 2018-01-11 | Chunghwa Telecom Co Ltd | Gender and age identification method for vertical image flow counting |
| CN108171162A (en) * | 2017-12-27 | 2018-06-15 | 重庆交通开投科技发展有限公司 | Crowded degree detection method, apparatus and system |
| CN108207653A (en) * | 2018-01-31 | 2018-06-29 | 佛山市神风航空科技有限公司 | A kind of grazing system based on unmanned plane |
| CN109461151A (en) * | 2018-11-05 | 2019-03-12 | 上海睿畜电子科技有限公司 | A kind of method, apparatus and system that livestock number is checked |
Non-Patent Citations (1)
| Title |
|---|
| 特征级图像融合及在医学图像中的应用研究;吴翠颖;周涛;陆惠玲;王媛媛;;电视技术(第12期);135-147 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110796632A (en) | 2020-02-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109815919B (en) | Crowd counting method, network, system and electronic equipment | |
| CN110889826B (en) | Eye OCT image focus region segmentation method, device and terminal equipment | |
| US11983950B2 (en) | Entity identification using machine learning | |
| WO2022033150A1 (en) | Image recognition method, apparatus, electronic device, and storage medium | |
| WO2020181872A1 (en) | Object detection method and apparatus, and electronic device | |
| JP7026456B2 (en) | Image processing device, learning device, focus control device, exposure control device, image processing method, learning method, and program | |
| CN105260750B (en) | A kind of milk cow recognition methods and system | |
| CN110796632B (en) | Pig counting device | |
| CN110263755B (en) | Eye ground image recognition model training method, eye ground image recognition method and eye ground image recognition device | |
| CN110781770B (en) | Living body detection method, device and equipment based on face recognition | |
| CN111178276B (en) | Image processing method, image processing apparatus, and computer-readable storage medium | |
| CN111401215B (en) | Multi-class target detection method and system | |
| CN108229351A (en) | A kind of animal-breeding method, animal-breeding device and electronic equipment | |
| CN113781421A (en) | Underwater-based target identification method, device and system | |
| JP7350208B2 (en) | Image processing device, image processing method, and program | |
| CN107678551A (en) | Gesture identification method and device, electronic equipment | |
| JP2016505185A (en) | Image processor having edge selection function | |
| CN111241928A (en) | Face recognition base optimization method, system, equipment and readable storage medium | |
| CN109767826A (en) | A kind of acquisition methods and medical photography imaging system of medical treatment photographed data | |
| CN112613471A (en) | Face living body detection method and device and computer readable storage medium | |
| JP6851246B2 (en) | Object detector | |
| CN111310532A (en) | Age identification method and device, electronic equipment and storage medium | |
| CN115081621A (en) | Model training method, focus segmentation device, computer device, and medium | |
| CN109741360B (en) | Bone joint segmentation method, device, terminal and readable medium | |
| JP4859057B2 (en) | Image processing apparatus, image processing method, program, and recording medium |
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 | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20230707 Address after: 401121 27F, building B, Pisces, No. 55, middle section of Huangshan Avenue, Yubei District, Chongqing Applicant after: Chongqing Yutonghe Digital Technology Co.,Ltd. Address before: 400050 floor 24, Yulong Building, No. 26-4, Yangjiaping Main Street, Jiulongpo District, Chongqing Applicant before: CHONGQING XIAOFUNONGKANG AGRICULTURAL TECHNOLOGY SERVICES Co.,Ltd. |
|
| TA01 | Transfer of patent application right | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |