CN215833929U - Fish character collection system based on machine vision technology - Google Patents
Fish character collection system based on machine vision technology Download PDFInfo
- Publication number
- CN215833929U CN215833929U CN202120933030.3U CN202120933030U CN215833929U CN 215833929 U CN215833929 U CN 215833929U CN 202120933030 U CN202120933030 U CN 202120933030U CN 215833929 U CN215833929 U CN 215833929U
- Authority
- CN
- China
- Prior art keywords
- tray
- fish
- boss
- machine vision
- vision technology
- 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
Images
Landscapes
- Length Measuring Devices By Optical Means (AREA)
Abstract
The utility model aims to provide a fish character acquisition device based on a machine vision technology, which is used for correcting the actual size represented by each pixel of images with different heights (fish body thickness) when acquiring fish body images by designing a high scale and a low scale so as to acquire more accurate character data such as fish body length, full length, body height, body thickness and the like. And (3) carrying out correction analysis on the two graduated scales with different heights on the acquired image, fitting a linear relation between the height and the actual distance of each pixel, and further obtaining the actual distance of each pixel at any height so as to obtain relatively more accurate character data.
Description
Technical Field
The utility model relates to the field of aquaculture measuring devices, in particular to a fish character acquisition device based on a machine vision technology.
Background
Machine vision measurement is a direction of development of measurement technology, for example, in aquatic science research, a machine vision technology is utilized to perform research such as measurement analysis and the like on fish body length, body height, body width, body surface characteristics and the like, when fish characters such as body length, body height, caudal peduncle length and the like are measured through the machine vision technology, a reference ruler is usually placed on a plane, a fish body is placed on the plane, a camera is used for vertically and downwards acquiring images of the fish body and the reference ruler to obtain a group image of the fish body and the reference ruler, pixels are converted through the reference ruler, so that various character parameters of the fish body are measured, generally, objects in the acquired image are small and large, if the fish body thickness is small, the reference ruler is closer to the surface of the fish body, the fish character parameters are measured by taking the reference ruler as a standard, and errors are within an acceptable range; when the fish body thickness is great, the reference scale plane is great with fish body surface distance, and the relative fish body surface distance camera of reference scale is far away to the pixel distance that this reference scale converted will increase the error when measuring fish characters, measures the failure.
The existing fish character acquisition platform does not have a platform integrating functions of fish body thickness, weight, image acquisition, chip scanning and the like, particularly, the acquired fish body image does not have image pixel correction data related to the fish body thickness, and the character data of fish body length, full length, body height, body thickness and the like cannot be accurately acquired through the fish body image.
SUMMERY OF THE UTILITY MODEL
In view of the problems in the prior art, an object of the present invention is to provide a fish character acquisition apparatus based on machine vision technology, which is designed with two scales for height, and is used to correct the actual size represented by each pixel of images with different heights (fish body thickness) when acquiring fish body images, so as to obtain more accurate character data of fish body length, full length, body height, body thickness, etc.
The object of the utility model is achieved in the following way:
a fish character acquisition device based on a machine vision technology comprises a tray, a camera and a computer; the method is characterized in that: the scale A and the scale B are arranged above the tray, and the scale A, the scale B and the upper plane of the tray are mutually parallel; the graduated scale A and the graduated scale B are different in height.
Two tip are provided with the boss of high difference around the tray, and the boss includes boss A and boss B, and boss A's last plane is provided with scale B, and the plane is close to boss A department on the tray and is provided with scale A, and the top of tray is higher than the position of boss and sets up the camera.
A weighing sensor is arranged below the tray, and a non-contact displacement sensor is arranged above the tray.
The tray is a transparent weighing tray, and the upper surface of the tray is a rectangular plane in the horizontal direction.
A backlight source is arranged below the tray, a shielding object is not arranged between the backlight source and the tray, and a positive light source is arranged above the tray and higher than the boss.
The boss is long-strip-shaped, and the height of the boss A is larger than that of the boss B.
The lens direction of the camera is vertically downward.
The measuring direction of the non-contact displacement sensor is vertically downward.
The computer is connected with a controller, a camera and a computer screen; the controller is electrically connected with a weighing sensor, a non-contact displacement sensor, a chip code scanner, a foot switch A, a foot switch B, a positive surface light source and a backlight source.
Compared with the prior art, the utility model has the following technical effects:
1. the device provided by the utility model integrates the functions of fish body thickness, weight, image acquisition, chip scanning and the like, measurement information is comprehensive and diversified, the measurement accuracy is higher, and more detailed basic data is provided for aquatic product research.
2. The device has high automation degree, reduces a large amount of manpower and material resources, and saves the measurement time.
3. And (3) carrying out correction analysis on the two graduated scales with different heights on the acquired image, fitting a linear relation between the height and the actual distance of each pixel, and further obtaining the actual distance of each pixel at any height so as to obtain relatively more accurate character data.
Drawings
FIG. 1 is a schematic view of the acquisition device of the present invention 1;
FIG. 2 is a schematic view of the collection device of the present invention 2;
FIG. 3 is a schematic view of the transparent weighing pan of the present invention;
FIG. 4 is a schematic view of the present invention;
wherein, 1, a camera; 2. a non-contact displacement sensor; 3. a front surface light source; 4. a transparent weighing pan; 5. a boss A; 6. a boss B; 7. a graduated scale A; 8. a graduated scale B; 9. a backlight source; 10. a weighing sensor; 11. a base; 12. a foot switch A; 13. a foot switch B; 14. a chip code scanner; 15. a computer; 16. and (5) measuring the thickness of the fish body.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings and specific embodiments.
As shown in fig. 1-4, a fish character collecting device based on machine vision technology comprises a tray, a camera 1 and a computer 15; the tray is characterized by further comprising a graduated scale A7 and a graduated scale B8 which are arranged above the tray, wherein the graduated scale A7, the graduated scale B8 and the upper plane of the tray are mutually parallel in pairs; the scale a7 and scale B8 are different in height.
Two tip in front and back of tray are provided with the boss of highly difference, and the boss includes boss A5 and boss B6, and boss A5's last plane is provided with scale B8, and the plane is provided with scale A7 near boss A5 department on the tray, and the position that the top of tray is higher than the boss sets up camera 1.
A weighing sensor 10 is arranged below the tray, and a non-contact displacement sensor 2 is arranged above the tray.
The tray is a transparent weighing tray 4, and the upper surface of the tray is a rectangular plane in the horizontal direction.
A backlight source 9 is arranged below the tray, no shielding object is arranged between the backlight source 9 and the tray, and a positive light source 3 is arranged above the tray and higher than the boss.
The difference in height between scale a and scale B is a definite value. The four sides under the transparent weighing plate can be provided with weighing sensor mounting holes, and the weighing sensors are arranged left and right or front and back to free the middle part for upward projection of backlight. Set up stand and cantilever as bearing supporting role in one side of base, the stand vertical fixation that makes progress on the base, the cantilever level is placed, connects on the stand, can follow the stand from top to bottom, back-and-forth movement, the one end fixed mounting camera lens of cantilever is perpendicular decurrent camera and the non-contact displacement sensor of measuring distance downwards, and the computer screen is installed to the cantilever below.
The lower end of the weighing sensor is fixed on the base, the upper end of the weighing sensor is fixedly connected with the left side or the right side or the front side and the back side below the transparent weighing plate, the backlight source is fixedly connected on the base and irradiates upwards, and the position of the backlight source is right below the transparent weighing plate. The whole device integrates the functions of fish body thickness, weight, image acquisition, chip scanning and the like, measurement information is comprehensive and diversified, the measurement accuracy is higher, and more detailed basic data are provided for aquatic product research; the device has high automation degree, reduces a large amount of manpower and material resources, and saves the measurement time. The actual distance of each pixel is calculated through analyzing two scales with different heights on the collected image by machine vision software, and then relatively more accurate character data on any height is obtained.
The bosses are long-strip-shaped, the height of the boss A5 is larger than that of the boss B6, the heights of the two bosses are not necessary, the height of the boss B is generally slightly lower, and the fish body placement is convenient to measure.
The direction of the lens of the camera 1 is vertical downwards, and the shooting surface of the camera lens is perpendicular to the upper surface of the transparent weighing tray, so that unnecessary errors in image acquisition are avoided.
The measuring direction of the non-contact displacement sensor 2 is vertical downward, and the laser ray is vertical to the upper surface of the transparent weighing tray, so that unnecessary errors of the measuring distance are avoided.
The computer 15 is connected with a controller, the camera 1 and a computer screen; the controller is electrically connected with a weighing sensor 10, a non-contact displacement sensor 2, a chip code scanner 14, a foot switch A12, a foot switch B13, a front light source 3 and a backlight source 9.
The device of the embodiment measures the process:
1. platform correction: (1) opening a backlight source and a front light source, moving the cantilever up and down and horizontally along the upright column to enable the camera view to be suitable for fish image acquisition, and fixing the cantilever by two scales on the transparent weighing plate in the camera view; (2) the non-contact displacement sensor collects the distance from the bottom surface of the transparent weighing plate to the non-contact displacement sensor at the moment, and records the value as a zero point; (3) the foot switch 1 is stepped on by feet to trigger the weighing sensor to reset; (4) the foot pedal switch 2 is stepped on by feet to trigger the camera to shoot, images of the graduated scale 1 and the graduated scale 2 are obtained, because the distance between the two graduated scales and the camera is different, the machine vision software carries out pixel correction on the two graduated scales in the images to obtain the actual distance value of each pixel, and according to the height difference of the two graduated scales, a method of a stepped machine vision measurement correction module in patent 201921459683.1 is adopted to obtain a linear formula of the actual distance value of each pixel, wherein the upward actual distance value of each pixel corresponds to the height by taking the bottom of a transparent weighing plate as a 0-point plane. The computer stores the zero value of the non-contact sensor, the zero value of the weighing sensor, the images of the two scales and a linear formula of the actual distance value of each pixel corresponding to the height with the bottom of the transparent weighing plate as a 0-point plane upward, and the correction is finished.
2. Platform collection: on the basis of correction, a chip code scanner is used for scanning a glass chip in a fish body to be measured to obtain the code of the fish, the side face of the fish body is upwards placed on a transparent weighing disc, a detection point (such as a light spot of a laser displacement sensor) of a non-contact displacement sensor is positioned at a fish body thickness measurement position (the highest point of the fish body, with the bottom of the transparent weighing disc being 0 point, in the vertical direction is generally easy to find according to other common methods of detection personnel), and the measured fish body thickness is the fish body thickness. And observing the fish body image, the weight and the body thickness data on a computer display screen, and when the parameter requirements of the height direction are met, stepping on the foot switch 2 by a foot to trigger the computer to acquire the image, and simultaneously acquiring the fish body weight data of the weighing sensor and the fish body thickness data of the non-contact displacement sensor, and removing the fish body from the transparent weighing plate after the acquisition is finished.
3. And (5) circulating the step 2, and acquiring the image, the body thickness and the weight data of each fish.
4. After the fish body is removed from the transparent weighing plate every time, weighing data can be checked, if the fish body does not return to zero, the foot switch 1 is stepped on by feet, the weighing sensor is triggered to be reset, and after all data parameters are reset, the image, the body thickness, the body weight and other property data of the fish body are continuously and circularly acquired.
5. And (3) processing the collected fish body images, body thickness data and weight data in batches through machine vision software to obtain the property data of the whole length, the body height, the body thickness, the weight and the like of each fish.
In the device, the camera and the non-contact displacement sensor can be replaced by a depth camera, the depth camera can collect a plane image of the fish body and can also collect height data of each point of the surface of the fish body from the bottom surface of the groove of the transparent weighing plate, and the character data of the whole length, the body height, the body thickness, the weight and the like of each fish can be obtained by the working method of 1-5.
Although the embodiments of the present invention have been described with reference to the accompanying drawings, it is not intended to limit the scope of the present invention, and it should be understood by those skilled in the art that various modifications and variations can be made without inventive efforts by those skilled in the art based on the technical solution of the present invention.
Claims (9)
1. A fish character acquisition device based on a machine vision technology comprises a tray, a camera (1) and a computer (15); the method is characterized in that: the tray is characterized by further comprising a graduated scale A (7) and a graduated scale B (8) which are arranged above the tray, wherein the graduated scale A (7), the graduated scale B (8) and the upper plane of the tray are parallel to each other; the graduated scale A (7) and the graduated scale B (8) are different in height.
2. The fish trait collection device based on machine vision technology of claim 1, wherein: two tip are provided with the boss of highly difference around the tray, and the boss includes boss A (5) and boss B (6), and the last plane of boss A (5) is provided with scale B (8), and the plane is close to boss A (5) department on the tray and is provided with scale A (7), and the position that the top of tray is higher than the boss sets up camera (1).
3. The fish trait collection device based on machine vision technology of claim 1, wherein: a weighing sensor (10) is arranged below the tray, and a non-contact displacement sensor (2) is arranged above the tray.
4. The fish trait collection device based on machine vision technology of claim 2, wherein: the tray is a transparent weighing tray (4), and the upper surface of the tray is a rectangular plane in the horizontal direction.
5. The fish trait collection device based on machine vision technology of claim 4, wherein: a backlight source (9) is arranged below the tray, a shielding object is not arranged between the backlight source (9) and the tray, and a front light source (3) is arranged above the tray and higher than the boss.
6. The fish trait collection device based on machine vision technology of claim 2, wherein: the boss is long-strip-shaped, and the height of the boss A (5) is greater than that of the boss B (6).
7. The fish trait collection device based on machine vision technology of claim 1, wherein: the lens direction of the camera (1) is vertically downward.
8. The fish trait collection device based on machine vision technology of claim 3, wherein: the measuring direction of the non-contact displacement sensor (2) is vertically downward.
9. The fish trait collection device based on machine vision technology of claim 1, wherein: the computer (15) is connected with a controller, a camera (1) and a computer screen; the controller is electrically connected with a weighing sensor (10), a non-contact displacement sensor (2), a chip code scanner (14), a foot switch A (12), a foot switch B (13), a front light source (3) and a backlight source (9).
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202120933030.3U CN215833929U (en) | 2021-04-30 | 2021-04-30 | Fish character collection system based on machine vision technology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202120933030.3U CN215833929U (en) | 2021-04-30 | 2021-04-30 | Fish character collection system based on machine vision technology |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN215833929U true CN215833929U (en) | 2022-02-15 |
Family
ID=80188472
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202120933030.3U Active CN215833929U (en) | 2021-04-30 | 2021-04-30 | Fish character collection system based on machine vision technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN215833929U (en) |
-
2021
- 2021-04-30 CN CN202120933030.3U patent/CN215833929U/en active Active
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108471725B (en) | Multifunctional livestock measuring station | |
| CA2001429A1 (en) | Pedotopography apparatus and method using moire fringe analysis to measure foot shapes | |
| CN208206040U (en) | A kind of circuit board electronic component height detection apparatus | |
| CN113048957A (en) | Vision-based rapid detection system for precision of laser demarcation device | |
| CN215833929U (en) | Fish character collection system based on machine vision technology | |
| CN207590657U (en) | A kind of intelligent physical examination all-in-one machine with laser spacing | |
| CN215812785U (en) | Slump measuring device based on image recognition and block chain | |
| CN109862351B (en) | Camera module resolution detection method | |
| CN208187381U (en) | Laser leveler automatic checkout system | |
| CN109357695A (en) | A kind of thermohygrometer automated detection system | |
| CN107389511B (en) | Crop seed phenotype measuring method and portable device | |
| CN213903347U (en) | Paster apparent on-line measuring equipment | |
| CN116164818A (en) | Determination method, device, equipment and storage medium for measuring uncertainty | |
| CN209689531U (en) | A kind of standardization external devices detection device | |
| CN113390754A (en) | Automatic verification/calibration device and method for glass float meter | |
| CN215525558U (en) | Non-woven fabric cloth cover style analyzer | |
| CN214952835U (en) | Automatic verification/calibration device for glass float meter | |
| CN205785454U (en) | Crop kernel mass of 1000 kernel determinator | |
| CN216979797U (en) | Efficient litchi fruit image acquisition platform | |
| CN114623781B (en) | Image-based measuring method and measuring system for outline dimensions of crop seeds such as green manure | |
| CN110411379A (en) | A kind of detection method of computer host box flatness detection | |
| RU2806064C1 (en) | Automated system for determining object dimensions | |
| RU2808941C1 (en) | Device and method for measuring overall dimensions of object | |
| CN211124646U (en) | Single slit diffraction experimental device for measuring refractive index of transparent medium | |
| CN114674252B (en) | Flatness detection system and method for protective middle frame of tablet personal computer |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| GR01 | Patent grant | ||
| GR01 | Patent grant |