A caged chicken individual tracking method based on multi-information fusion

Abstract

The present invention discloses a method for tracking individual caged chickens based on multi-information fusion. The method comprises the following steps: A. installing an RFID receiver on the top of the cage and two image acquisition sensors on the side walls of the cage, with the two image acquisition sensors shooting at a 90° angle. The chickens in the cage wear leg rings with RFID chips; B. the two image acquisition sensors synchronously capture the positions of the chickens, and the two captured images are used to perform preliminary positioning of the chickens; C. the RFID receiver captures RFID chip signals during each image capture, and determines the distance between each chicken and the RFID receiver based on the captured RFID chip signals; D. using the preliminary positioning information of the chickens and the distance between each chicken and the RFID receiver to determine the precise location of each chicken. The present invention can improve the shortcomings of existing technologies and simplify the process of tracking individual chickens.

Description

Technical Field

[0001] The present invention relates to the technical field of caged chickens, and in particular to a method for tracking individual caged chickens based on multi-information fusion. Background Art

[0002] Cage-based livestock and poultry farming offers a high-density, low-cost method. However, due to insufficient space, this can lead to poor growth and development. This necessitates accurate tracking of individual livestock and poultry movements to assess their activity levels and health. While RFID technology alone is a relatively mature method for locating and tracking individual livestock, it imposes significant hardware limitations and cumbersome data processing. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide a caged chicken individual tracking method based on multi-information fusion, which can solve the shortcomings of the existing technology and simplify the tracking process of individual chickens.

[0004] In order to solve the above technical problems, the technical solutions adopted by the present invention are as follows.

[0005] A caged chicken individual tracking method based on multi-information fusion includes the following steps:

[0006] A. Install an RFID receiver on the top of the chicken coop and two image acquisition sensors on the side walls of the chicken coop. The shooting directions of the two image acquisition sensors are at an angle of 90°. The chickens in the chicken coop wear leg rings with RFID chips.

[0007] B. Two image acquisition sensors synchronously capture the position of the chickens and use the two captured images to perform preliminary positioning of the chickens;

[0008] C. The RFID receiver collects RFID chip signals at the same time as each image acquisition sensor takes a picture, and determines the distance between different chickens and the RFID receiver based on the collected RFID chip signals;

[0009] D. Use the chicken's preliminary positioning information and the distance between different chickens and the RFID receiver to determine the precise location of each chicken.

[0010] Preferably, in step B, using the two captured images to perform preliminary positioning of the chickens includes the following steps:

[0011] B1. Divide the two images into blocks according to the shooting positions, and then group two image blocks with the same shooting position but different shooting angles into one group, to obtain several image groups;

[0012] B2. Identify the chickens in each image group;

[0013] B3. Combine the identified chickens with the shooting positions of the image blocks to obtain preliminary positioning information of the chickens.

[0014] Preferably, in step B1, an image block grayscale variance threshold is set, and when image segmentation is performed, the grayscale variances of all image blocks are made greater than the image block grayscale variance threshold.

[0015] Preferably, in step B2, chicken identification includes the following steps:

[0016] B21, mark the chicken feature points that appear simultaneously in two image blocks in the image group;

[0017] B22, mark the chicken feature points that only appear in one image block;

[0018] B23, using the chicken feature points marked in step B21 to restore the individual chickens;

[0019] B24. Mark the restored chicken in the image block, then compare the chicken feature points marked in step B22 with the restored chicken to filter out the chicken feature points that do not belong to the restored chicken.

[0020] B25. Use the chicken feature points selected in step B24 to restore the individual chickens;

[0021] B26. The individual chickens restored in steps B23 and B25 are regarded as identified chickens.

[0022] Preferably, in step D, determining the precise location of each chicken comprises the following steps:

[0023] Based on the signal received by the RFID receiver, the distance between the signal and the RFID receiver is determined, and the possible circular position range of the signal is drawn with this distance as the radius. The preliminary positioning information of the chicken is compared with the drawn circular position range to determine the precise position of each chicken.

[0024] Preferably, in step D, when preliminary positioning information of multiple chickens appears in the circular position range, the preliminary positioning information of the multiple chickens is screened according to the historical trajectory of the signal corresponding to the circular position range.

[0025] The beneficial effects of the above technical solution are: by combining visual recognition with RFID ranging, the present invention uses visual recognition to quickly identify chickens, and then combines it with RFID ranging data to obtain the chicken's accurate location. The hardware layout is highly flexible and the data calculation process is simple. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 It is a schematic diagram of the present invention. DETAILED DESCRIPTION

[0027] A specific embodiment of the present invention includes the following steps:

[0028] A. Install an RFID receiver on the top of the chicken coop and two image acquisition sensors on the side walls of the chicken coop. The shooting directions of the two image acquisition sensors are at an angle of 90°. The chickens in the chicken coop wear leg rings with RFID chips.

[0029] B. Two image acquisition sensors synchronously capture the position of the chickens, and use the two captured images to perform preliminary positioning of the chickens. The preliminary positioning of the chickens using the two captured images includes the following steps:

[0030] B1. Divide the two images into blocks according to the shooting positions, and then group two image blocks with the same shooting position but different shooting angles into a group to obtain several image groups; set the image block grayscale variance threshold so that the grayscale variance of all image blocks is greater than the image block grayscale variance threshold when dividing the images into blocks;

[0031] B2. Identify the chickens in each image group. Chicken identification includes the following steps:

[0032] B21, mark the chicken feature points that appear simultaneously in two image blocks in the image group;

[0033] B22, mark the chicken feature points that only appear in one image block;

[0034] B23, using the chicken feature points marked in step B21 to restore the individual chickens;

[0035] B24. Mark the restored chicken in the image block, then compare the chicken feature points marked in step B22 with the restored chicken to filter out the chicken feature points that do not belong to the restored chicken.

[0036] B25. Use the chicken feature points selected in step B24 to restore the individual chickens;

[0037] B26, taking the individual chickens restored in steps B23 and B25 as identified chickens;

[0038] B3. Combine the identified chickens with the shooting positions of the image blocks to obtain preliminary positioning information of the chickens.

[0039] C. The RFID receiver collects RFID chip signals at the same time as each image acquisition sensor takes a picture, and determines the distance between different chickens and the RFID receiver based on the collected RFID chip signals;

[0040] D. Determine the precise location of each chicken using the chicken's preliminary location information and the distance between each chicken and the RFID receiver; determining the precise location of each chicken includes the following steps:

[0041] Based on the signal received by the RFID receiver, the distance between the signal and the RFID receiver is determined, and the possible circular position range of the signal is drawn with this distance as the radius. The preliminary positioning information of the chicken is compared with the drawn circular position range to determine the precise position of each chicken. When the preliminary positioning information of multiple chickens appears in the circular position range, the preliminary positioning information of multiple chickens is screened according to the historical trajectory of the corresponding signals in the circular position range.

[0042] In the description of the present invention, it should be understood that the terms "longitudinal", "transverse", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating the orientation or position relationship, are based on the orientation or position relationship shown in the accompanying drawings, and are only for the convenience of describing the present invention, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention.

[0043] The basic principles, main features, and advantages of the present invention are shown and described above. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions are merely illustrative of the principles of the present invention. Various changes and modifications may be made to the present invention without departing from the spirit and scope of the present invention. Such changes and modifications are intended to fall within the scope of the present invention. The scope of protection claimed in the present invention is defined by the appended claims and their equivalents.

Claims

1. A caged chicken individual tracking method based on multi-information fusion, characterized in that The following steps are involved: A. Install an RFID receiver on the top of the chicken coop and two image acquisition sensors on the side walls of the chicken coop. The shooting directions of the two image acquisition sensors are at an angle of 90°. The chickens in the chicken coop wear leg rings with RFID chips. B. Two image acquisition sensors synchronously capture the position of the chickens and use the two captured images to perform preliminary positioning of the chickens; C. The RFID receiver collects RFID chip signals at the same time as each image acquisition sensor takes a picture, and determines the distance between different chickens and the RFID receiver based on the collected RFID chip signals; D. Use the chicken's preliminary positioning information and the distance between different chickens and the RFID receiver to determine the precise location of each chicken; In step B, the preliminary positioning of the chicken using the two captured images includes the following steps: B1. Divide the two images into blocks according to the shooting positions, and then group two image blocks with the same shooting position but different shooting angles into one group, to obtain several image groups; B2. Identify the chickens in each image group; B3, combining the identified chickens with the shooting positions of the image blocks to obtain preliminary positioning information of the chickens; In step B2, chicken identification includes the following steps: B21, mark the chicken feature points that appear simultaneously in two image blocks in the image group; B22, mark the chicken feature points that only appear in one image block; B23, using the chicken feature points marked in step B21 to restore the individual chickens; B24. Mark the restored chicken in the image block, then compare the chicken feature points marked in step B22 with the restored chicken to filter out the chicken feature points that do not belong to the restored chicken. B25. Use the chicken feature points selected in step B24 to restore the individual chickens; B26. The chicken individuals restored in steps B23 and B25 are regarded as identified chickens.

2. The caged chicken individual tracking method based on multi-information fusion according to claim 1 is characterized in that: In step B1, a grayscale variance threshold of an image block is set, and when image segmentation is performed, the grayscale variances of all image blocks are made to be greater than the grayscale variance threshold.

3. The caged chicken individual tracking method based on multi-information fusion according to claim 1 is characterized in that: In step D, determining the exact location of each chicken includes the following steps: Based on the signal received by the RFID receiver, the distance between the signal and the RFID receiver is determined, and the possible circular position range of the signal is drawn with this distance as the radius. The preliminary positioning information of the chicken is compared with the drawn circular position range to determine the precise position of each chicken.

4. The caged chicken individual tracking method based on multi-information fusion according to claim 3 is characterized in that: In step D, when preliminary positioning information of multiple chickens appears in the circular position range, the preliminary positioning information of the multiple chickens is screened according to the historical trajectory of the signal corresponding to the circular position range.

Images