Egg crack detection equipment, system and method based on machine vision

Abstract

The invention discloses an egg crack detection device, system and method based on machine vision, and relates to the technical field of machine vision. The image acquisition device is arranged on the conveying line and is used for acquiring egg images; and the image processing device is connected with the image acquisition device and is used for detecting whether crack characteristics exist on the surfaces of the eggs or not according to the acquired egg images in combination with a deep learning algorithm and grading the eggs according to crack detection states of the eggs, so that automatic identification of non-contact crack characteristics of the eggs and automatic grading of the eggs are realized, and the processing efficiency of the eggs is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of machine vision, in particular to an egg crack detection device, system and method based on machine vision. BACKGROUND

[0002] In recent years, with the continuous increase of domestic labor cost, poultry hatching and breeding enterprises as labor-intensive enterprises have to face increasing labor cost and management cost pressure. At present, the daily egg handling capacity of a medium-sized hatching farm is about 200,000, and the egg handling process is basically completed by manual work, which has high labor intensity, slow processing speed, poor processing effect, and brings a series of adverse factors to the subsequent hatching process, and also greatly restricts the scale development of the hatching farm. The existing equipment in the market can realize automatic detection of egg cracks through acoustic detection, but the detection cost is high, and the detection adopts a contact type knocking method, which is easy to cause secondary damage to the eggs and has the risk of cross contamination, so it is difficult to meet the needs of large-scale hatching farms.

[0003] Therefore, the present application provides an egg crack detection device, system and method based on machine vision. SUMMARY

[0004] The present application provides an egg crack detection device, system and method based on machine vision, which acquires egg images by setting an image acquisition device, analyzes the egg images to obtain the crack state of the egg, and realizes automatic identification of the egg crack through a fixed algorithm, thereby improving the processing efficiency of the poultry eggs and improving the labor environment of the workers.

[0005] According to an aspect of the present disclosure, an egg crack detection device based on machine vision is provided, characterized in that the detection device comprises: a conveying line 1 for conveying eggs; an image acquisition device 2 arranged on the conveying line 1 for acquiring egg images; an image processing device connected with the image acquisition device, which automatically identifies the egg surface crack according to the acquired egg images combined with a deep learning algorithm, and judges whether the egg has a crack area.

[0006] In a possible implementation, the image processing device performs segmentation and enhancement preprocessing operations on the original egg image according to the original egg image, and determines the egg crack detection state combined with a deep learning model.

[0007] In a possible implementation, the side of the conveying line 1 is provided with a light source device 3, and the light source device 3 comprises a light source 31 and a light source fixing bracket 32. The light source 31 is used to provide illumination light for the eggs, and the light source fixing bracket 32 is used to fix the light source 31. The light source 31 is installed on the conveying line 1 through a light source fixing support 32, and the light source 31 is located above or below the conveying line 1.

[0008] In a possible implementation, the image acquisition device 2 comprises one or more industrial cameras 21, a fixing support 22 of the industrial camera 21 and a shielding cover 23. The industrial camera 21 is located above the conveying line 1, and the industrial camera 21 is horizontally arranged on the conveying line 1 through the fixing support 22; the industrial camera 21 and the light source 31 form an egg detection channel therebetween; The shielding cover 23 is arranged around and on the top of the fixing support 22, the inside of the shielding cover 23 is coated with a black coating, and a conveying channel is formed between the lower part of the shielding cover 23 and the conveying line 1. The industrial camera 21 and the light source 31 are located in the conveying channel, and a protective rubber strip is arranged at the entrance and exit of the conveying channel formed between the shielding cover 23 and the conveying line 1.

[0009] The egg crack detection system based on machine vision comprises: The egg crack detection device based on machine vision and an egg grader connected with the detection device; The egg grader grades the eggs according to the detection result of the egg crack detection device.

[0010] The egg crack detection method based on machine vision is used for the detection system, and the method comprises: Acquiring an original image of the egg; Segmenting the original image according to the original image of the egg; Enhancing the image according to the segmented image of the egg to improve the contrast of the crack feature region; Realizing automatic detection of the egg crack according to the preprocessed image and in combination with a deep learning network; Grading the egg according to the crack detection state of the egg.

[0011] Compared with the prior art, the beneficial effects of the present application are: The egg crack recognition device based on machine vision disclosed in the embodiment automatically conveys the eggs to the image acquisition device through the conveying line, irradiates the eggs by the light source after the eggs enter the image acquisition device, photographs the eggs by the industrial camera, acquires the egg images, sends the acquired egg images to the image processing device for processing, detects whether there is a crack feature on the surface of the egg in combination with a deep learning algorithm, grades the egg according to the crack detection state of the egg, realizes automatic recognition of the non-contact crack feature of the egg and automatic grading of the egg, improves the processing efficiency of the poultry eggs and improves the labor environment of workers.

[0012] By setting the image acquisition device to acquire egg images, the egg images are analyzed and combined with a deep learning network to realize automatic identification of the non-contact crack of the egg, improve the processing efficiency of the poultry egg, and improve the labor environment of the workers.

[0013] The present disclosure is matched with an egg grader, and the egg is automatically graded by the crack detection state of the egg, further improving the processing efficiency and quality of the poultry egg and improving the labor environment of the workers. BRIEF DESCRIPTION OF DRAWINGS

[0014] Figure 1 It is a structural schematic diagram of an embodiment of the present disclosure.

[0015] Figure 2 It is a structural schematic diagram of an image acquisition device in an embodiment of the present disclosure.

[0016] Figure 3 It is a structural schematic diagram of a light source device in an embodiment of the present disclosure.

[0017] Figure 4 It is an effect picture of a breeder egg collected in an embodiment of the present disclosure.

[0018] Figure 5 It is an effect picture of a breeder egg processed in an embodiment of the present disclosure.

[0019] Figure 6 It is a flow chart of egg crack detection in an embodiment of the present disclosure.

[0020] Wherein: 1, conveying line, 2, image acquisition device, 3, light source device, 21, industrial camera, 22, fixed support; 23, shielding cover; 31, light source, 32, light source fixed support. DETAILED DESCRIPTION

[0021] Various exemplary embodiments, features and aspects of the present disclosure will be described in detail below with reference to the accompanying drawings. The same reference signs in the drawings represent functionally the same or similar elements. Although various aspects of the embodiments are shown in the drawings, the drawings are not necessarily drawn to scale unless specifically indicated.

[0022] The word "exemplary" is used herein in the sense of being an example, rather than a preference or requirement. Any embodiment described as "exemplary" is not necessarily to be construed as being preferred or advantageous over other embodiments.

[0023] In addition, in order to better illustrate the present disclosure, a large number of specific details are given in the specific embodiments below. Those skilled in the art should understand that the present disclosure can also be implemented without certain specific details. In some examples, methods, means, elements and circuits well known to those skilled in the art are not described in detail, in order to highlight the main idea of the present disclosure.

[0024] Example 1: In this example, an egg crack detection device based on machine vision is disclosed, the structure of which is as follows: Figure 1 、 Figure 2 and Figure 3 Shown, including: Conveyor line 1, used for automatically conveying hatching eggs; An image acquisition device 2 is provided on the conveyor line 1 and is used to acquire images of eggs; The image processing device is connected to the image acquisition device 2, receives the egg image acquired by the image acquisition device 2, analyzes the egg image, and determines whether there is a crack area in the egg in combination with a deep learning algorithm.

[0025] Image acquisition device 2 includes an industrial camera 21 and a light source device 3. Light source device 3 includes a light source 31 for illuminating the eggs and a light source mounting bracket 32 ​​for securing light source 31. Industrial camera 21 is positioned above conveyor line 1 and spans across conveyor line 1 via mounting bracket 22. Light source 31 is mounted on conveyor line 1 via mounting bracket 32 ​​and is positioned above or below conveyor line 1. An egg inspection channel is formed between industrial camera 21 and light source 31.

[0026] In this embodiment, the fixing bracket 22 can be obtained by connecting multiple channel steels, and includes at least one beam for fixing the industrial camera; the specific fixing and connection methods are conventional technologies and will not be described in detail here.

[0027] In this embodiment, the light source fixing bracket 32 ​​includes at least two vertical plates, and multiple horizontal plates for fixing the light sources are fixed between the two vertical plates. It can be understood that multiple light source fixing holes are provided on the horizontal plates; the two ends of the horizontal plates can be connected to the vertical plates by providing a threaded rotating shaft, and a fastening nut can be provided on the rotating shaft, and the angle of the horizontal plate can be adjusted and fixed by adjusting the tightness of the fastening nut; in other embodiments, the angle of the horizontal plate can be adjusted by selecting other methods, as long as the angle of one plate can be adjusted on its fixed object; the angle of the horizontal plate is adjustable, which realizes the irradiation of eggs at different angles, improves the flexibility of use, and achieves the purpose of irradiating eggs at multiple angles; at the same time, the multiple horizontal plates can also achieve cross-overlapping of light emitted by different light sources by adjusting different angles. When the cross-overlapping light is irradiated on the same egg, the illumination of the egg can be improved, and the key detection of eggs at different positions can be improved.

[0028] In order to reduce the influence of ambient light on the egg image acquisition, a shielding cover 23 is arranged on the image acquisition device 2, the shielding cover 23 is arranged around and on the top of the fixed support 22, the inside of the shielding cover 23 is coated with a black coating, a conveying channel is formed between the lower part of the shielding cover 23 and the conveying line 1, the industrial camera 21 and the light source 31 are located in the conveying channel, and a protective rubber strip is arranged at the entrance and exit of the conveying channel formed between the shielding cover 23 and the conveying line 1, so that the eggs can be conveniently conveyed in and out while the influence of ambient light is eliminated.

[0029] The image processing device receives the original image of the egg collected by the image acquisition device, performs segmentation on the original image according to the original image of the egg, performs an enhancement operation on the image according to the segmented image of the egg, improves the contrast of the crack feature area, realizes automatic detection of the egg crack according to the preprocessed image combined with a deep learning network, and performs grading processing on the egg according to the crack detection state of the egg.

[0030] The image processing device is connected with a PLC or other controller.

[0031] The egg crack recognition device based on machine vision disclosed in the embodiment automatically conveys the eggs to the image acquisition device through the conveying line, irradiates the eggs by the light source after the eggs enter the image acquisition device, takes photos of the eggs by the industrial camera, acquires the egg images, sends the collected egg images to the image processing device for processing, detects whether there is a crack feature on the surface of the egg combined with a deep learning algorithm, grades the eggs according to the crack detection state of the eggs, realizes automatic recognition of the non-contact crack feature of the eggs and automatic grading of the eggs, improves the processing efficiency of the poultry eggs, and improves the labor environment of workers.

[0032] In this embodiment, an egg crack recognition system based on machine vision is disclosed, which comprises the egg crack recognition device based on machine vision in embodiment 1 and an egg grader connected with the recognition device.

[0033] The egg grader grades the eggs according to the crack determination result of the eggs.

[0034] As shown in Figure 6 When the system is running, the stick line is running, it is judged whether the indexing plate detection is completed, if yes, the crack light source is triggered, the camera is triggered to take photos to acquire the original image of the egg, the original image is segmented, the image is enhanced to improve the contrast of the crack feature area, the image enhanced picture is input into the crack detection model to obtain a detection result, and the eggs are graded according to the detection result.

[0035] In this embodiment, an egg crack detection method based on machine vision is disclosed, which comprises: collecting an original image of the egg; segmenting the original image according to the original image of the egg; enhancing the image according to the segmented image of the egg to improve the contrast of the crack feature region; realizing automatic detection of the egg crack according to the preprocessed image combined with a deep learning network; grading the egg according to the crack detection state of the egg.

[0036] The above has described various embodiments of the present disclosure, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and changes are obvious to those skilled in the art without departing from the scope and spirit of the described embodiments. The selection of the terms used herein is intended to best explain the principles, practical applications, or technical improvements in the market of the embodiments, or to enable other ordinary skilled persons in the art to understand the embodiments disclosed herein.

Claims

1. Egg crack detection equipment based on machine vision, characterized in that: The detection equipment includes: A conveyor line (1) for conveying eggs; An image acquisition device (2) is provided on the conveyor line (1) and is used to acquire images of eggs; The image processing device is connected to the image acquisition device, and automatically identifies cracks on the egg surface based on the collected egg images in combination with a deep learning algorithm to determine whether there are crack areas on the egg.

2. The egg crack detection device based on machine vision according to claim 1, characterized in that: The image processing device performs segmentation and enhancement preprocessing operations on the original egg image based on the original egg image, and combines the deep learning model to determine the crack detection status of the breeding egg.

3. The egg crack detection device based on machine vision according to claim 1, characterized in that: A light source device (3) is provided on the side of the conveyor line (1), and the light source device (3) comprises a light source (31) and a light source fixing bracket (32); The light source (31) is used to provide irradiation light for the eggs, and the light source fixing bracket (32) is used to fix the light source (31); The light source (31) is mounted on the conveyor line (1) via a light source fixing bracket (32), and the light source (31) is located above or below the conveyor line (1).

4. The egg crack detection device based on machine vision according to claim 2, characterized in that: The image acquisition device (2) comprises: one or more industrial cameras (21), a fixing bracket (22) and a shielding cover (23) of the industrial camera (21); The industrial camera (21) is located above the conveyor line (1), and the industrial camera (21) is positioned across the conveyor line (1) via a fixed bracket (22); an egg detection channel is formed between the industrial camera (21) and the light source (31); The shielding cover (23) is arranged around and on the top of the fixed bracket (22), the interior of the shielding cover (23) is coated with a black coating, and a conveying channel is formed between the lower part of the shielding cover (23) and the conveying line (1); The industrial camera (21) and the light source (31) are located in the conveying channel, and protective rubber strips are provided at the entrance and exit of the conveying channel formed between the shielding cover (23) and the conveying line (1).

5. The egg crack detection system based on machine vision is characterized by: The system comprises: The machine vision-based egg crack detection device according to any one of claims 1 to 4, and an egg grader connected to the detection device; The egg grader grades eggs according to the test results of the egg crack detection equipment.

6. The egg crack detection method based on machine vision is characterized in that: The method is used in the detection system according to claim 5, and the method comprises: Collect original images of eggs; According to the original image of the egg, the original image is segmented; Based on the segmented image of the egg, the image is enhanced to improve the contrast of the crack feature area; Based on the pre-processed images, combined with the deep learning network, automatic detection of egg cracks is achieved; Eggs are graded based on their crack detection status.

Images