CN108655021B - Online damaged discernment sorter of maize seed based on machine vision - Google Patents

Abstract

The invention relates to the technical field of machine vision detection, in particular to an online damage recognition and classification device for corn seeds based on machine vision. Including feeding device, conveying device, image acquisition device, sorting device, control system, fixed-axis test bench, and recycling seed box. The characteristics are as follows: the corn seed online damage identification and classification device can collect dynamic images and realize the dynamic detection function of corn seed online damage identification, so as to separate the damaged corn seeds from the qualified corn seeds; a rotary disc transmission mechanism with a simple structure is designed; An unqualified corn seed removal mechanism is designed, which can accurately and effectively remove unqualified corn seeds; an online damage identification and classification control system for corn seeds is designed; the invention has simple structure, accurate identification and classification, high efficiency and good effect. The fully automated online real-time detection of corn seeds can make the detection of agricultural products more mechanized, automated and intelligent.

Description

An online damage recognition and classification device for corn seeds based on machine vision

technical field

The invention relates to the technical field of machine vision detection, in particular to an online damage recognition and classification device for corn seeds based on machine vision.

Background technique

As one of the important food crops in my country, corn has become the largest food crop in my country since 2012, and has a very important strategic position. Because corn seeds will be damaged by various factors such as mechanical gnawing during the process of harvesting and storage, which directly affects the germination rate of corn seeds and corn yield, the damage detection of corn seeds before sowing is essential. This process can ensure the germination rate of corn seeds, which is of great significance to the increase of corn yield.

Compared with manual detection technology, machine vision detection technology has the advantages of high speed, high precision, and good repeatability, and has a wide range of applications in the field of non-destructive testing of agricultural product quality. The use of machine vision detection technology to replace manual detection is an inevitable trend in the development of automatic classification. The research on grain appearance quality detection technology based on machine vision has high theoretical value and practical significance. At present, domestic research on maize seeds based on machine vision mainly focuses on the quality identification and grading of maize seeds, and mainly focuses on static detection, and there is no highly commercialized equipment for maize seed damage detection. The use of machine vision technology to detect and study the quality of grain appearance is of great significance to improving the market competitiveness of my country's grain.

SUMMARY OF THE INVENTION

The purpose of the present invention is to provide an online damage identification and classification device for corn seeds based on machine vision, so as to realize automatic online detection of corn seeds and to separate damaged corn seeds from the detected corn seeds.

An online damage recognition and classification device for corn seeds based on machine vision, which mainly includes: a feeding device [1], a conveying device [2], an image acquisition device [3], a classification device [4], a control system [5], Fixed axis test bench [6], recycling seed box [7].

The feeding device is composed of a finger clip-type seed metering device [8], a seed guiding tube [11], an infrared radiation sensor [10], a transmission gear set [12], a stepping motor a and its driver [14], It consists of a seat frame [9] and a seed storage box [13]. The finger clip-type seed metering device is installed vertically on the seat frame, the seed guiding tube is installed at the seed outlet of the seed metering device, the infrared radiation sensor is installed at both ends of the seed guiding tube outlet, and the two gears of the transmission gear set are respectively connected to the stepper motor. The output shaft and the intermediate shaft of the finger-clip type seed metering device cooperate with the transmission.

The conveying device is composed of a transmission turntable [14], a transmission gear set [15], a stepping motor b and its driver [16]. The transmission turntable is installed on the fixed-axis test bench, and the middle of the transmission turntable and the fixed-axis test bench is connected with a bearing. The driving wheel of the gear group is installed on the output shaft of the stepper motor, and the driven wheel is glued to the transmission turntable and concentric with the transmission turntable. coaxial. The power is output through the stepping motor, and the gear transmission is used to reduce the speed and increase the torque to drive the transmission turntable to rotate.

The image acquisition device is composed of an LED ring light source [17], a camera [18], and a connecting screw [19]. The LED ring light source is installed on the camera lens and is coaxial with the optical axis of the camera. The optical axis of the camera is perpendicular to the transmission turntable of the conveying device. The connecting screw fixes the camera on the fixed-axis experimental bench.

The classification device is divided into two mechanisms, a broken seed removal mechanism [20] and a qualified seed recovery mechanism [21], which are specifically composed of a seed drop groove [22], an opening and closing bottom cover [23], a closing It consists of a page, a magnet, a stepping motor c[24], a lever, and a connecting screw [25]. There are 8 circular holes on the transfer turntable of the conveying device, and the included angle between the centers of each two adjacent holes is 45°. The seed dropping groove is connected with the opening and closing bottom cover, and the opening and closing bottom cover is installed on the lower side of the seed dropping groove through hinges. The two magnets are respectively installed at the opening end of the opening and closing bottom cover and the bottom end of the seed-dropping groove, and the included angle with the hinge is 180°. The connecting screw is connected to the fixed axis of the fixed axis test bench. The horizontal installation angle between the feeding device and the image capture device is 45°, the horizontal installation angle between the image capture device and the damaged seed removal mechanism of the classification device is 90°, and the opening of the damaged seed removal mechanism of the classification device and the qualified seed recovery mechanism The horizontal installation angle of the cover lever is 90°, and the horizontal installation angle of the cover opening lever and the closing lever of the qualified seed recovery mechanism is 90°.

The control system includes: a computer, a single-chip microcomputer, an image acquisition and processing system, a stepper motor drive hardware part, a power control system software part, and the image detection judgment result is transmitted to the power control system in real time.

The fixed-axis experimental bench is an experimental bench with a shaft diameter of 10 mm and a weight of 2.5 kg.

The recovery seed box includes a qualified seed recovery seed box and a damaged corn seed recovery seed box, which are respectively placed under the opening and closing bottom cover of the broken seed rejection position and the opening and closing bottom cover of the unified recovery position of qualified seeds.

The beneficial effects of the present invention are:

(1) The present invention is based on machine vision detection technology, and is designed for the problem of damage of corn seeds due to various factors during growth, harvesting, storage and other processes, and can effectively separate the damaged corn seeds from the detected corn seeds. Come out, to achieve the purpose of corn seed selection. (2) It can realize fully automated online real-time detection of corn seeds, without manual labor, making the detection of agricultural products more mechanized, automated and intelligent.

Description of drawings

FIG. 1 is a schematic structural diagram of an online damage recognition and classification device for corn seeds based on machine vision.

Fig. 2 is a feeding device of a maize seed online damage recognition and classification device based on machine vision.

Fig. 3 is a conveying device of a maize seed online damage identification and classification device based on machine vision.

Fig. 4 is an image acquisition device of an online damage recognition and classification device for corn seeds based on machine vision.

Fig. 5 is a classification device of a maize seed online damage recognition and classification device based on machine vision.

Detailed ways

The present invention will be further described in detail below with reference to the embodiments given in the accompanying drawings.

Referring to FIG. 1, an online damage recognition and classification device for corn seeds based on machine vision mainly includes a feeding device [1], a conveying device [2], an image acquisition device [3], a classification device [4], and a control system [5] ], fixed axis test bench [6], recycling seed box [7].

Referring to Fig. 2 , a feeding device for an online damage recognition and classification device for corn seeds based on machine vision mainly includes a finger-clip-type seed metering device [8], a seed guiding tube [11], an infrared beam sensor [10], a transmission Gear set [12], stepping motor a and its driver [14], seat frame [9], seed storage box [13].

Referring to FIG. 3 , a conveying device of a maize seed online damage identification and classification device based on machine vision. It mainly includes: transmission turntable [14], transmission gear set [15], stepper motor b and its driver [16].

Referring to FIG. 4 , an image acquisition device for an online damage recognition and classification device for corn seeds based on machine vision mainly includes: an LED ring light source [17], a camera [18], and a connecting screw [19].

Referring to Fig. 5, a classification device based on a machine vision-based online damage recognition and classification device for corn seeds mainly includes two mechanisms: a damaged seed removal mechanism [20] and a qualified seed recovery mechanism [21], which are specifically composed of the seed drop groove [20]. 22], an opening and closing bottom cover [23], a hinge, a magnet, a stepping motor c [24], a lever, and a connecting screw [25].

When working: the stepper motor a of the feeding part drives the seed meter to perform periodic intermittent single seed metering, and the stepper motor b drives the transmission turntable to perform clockwise intermittent motion. Through the seed guide tube, it falls smoothly into the seed drop groove. When the corn seeds pass through the seed guide tube, an external interruption of the infrared beam sensor is triggered, and the signal is transmitted to the upper computer. The stepping motor b drives the transfer turntable to rotate 45° to transport the corn seeds. To the image acquisition device, the host computer controls the camera to collect the corn seed image, and sends the collected image back to the host computer. After the host computer judges the eligibility of the corn seed, if the corn seed is judged to be damaged, the host computer controls the step The motor c drives the lever to rotate 90°, and opens the opening and closing bottom cover, so that the damaged corn seeds fall into the recycling seed box below, thereby completing the removal of the damaged corn seeds. The qualified corn will rotate 180° with the transfer turntable to the qualified seed recovery mechanism. The bottom cover will be opened by the fixed lever. The qualified seeds will fall into the qualified seed recovery seed box below, and the transfer turntable will continue to rotate 90° clockwise. At the next fixed lever position, the open and closed bottom covers in the open state are closed uniformly by the lever. The transfer carousel continues to rotate clockwise to enter the next detection cycle.

Claims (5)

1. The utility model provides an online damaged discernment sorter of maize seed based on machine vision, characterized by: comprises a feeding device (1), a conveying device (2), an image acquisition device (3), a classification device (4), a control system (5), a dead axle experiment table (6) and a recovery seed box (7);

the feeding device consists of a finger-clamping seed sowing device (8), a seed guide tube (11), an infrared correlation sensor (10), a transmission gear set (12), a stepping motor a and a drive (14) thereof, a seat frame (9) and a seed storage box (13); the finger-clamping type seed metering device is vertically arranged on the seat frame, the seed guide tube is arranged at the seed outlet of the seed metering device, the infrared correlation sensors are arranged at two ends of the seed guide tube outlet, the driving wheel of the transmission gear set is connected with the output shaft of the stepping motor, the driven wheel is connected with the intermediate shaft of the finger-clamping type seed metering device, and the two gears are in matched transmission; the method is characterized in that: detecting whether seeds pass through by using an infrared correlation sensor, and finishing the rotation seed metering of the finger-clamp type seed metering device by matching a gear set with a stepping motor;

the conveying device consists of a conveying turntable (14), a transmission gear set (15), a stepping motor b and a driver (16) thereof; the method is characterized in that: the transmission turntable is arranged on the fixed-shaft experiment table, the transmission turntable is connected with the middle of the fixed-shaft experiment table through a bearing, a driving wheel of the gear set is arranged on an output shaft of the stepping motor, and a driven wheel is arranged on the transmission turntable and is concentric and coaxial with the transmission turntable; the power is output by the stepping motor, and the transmission speed is reduced and the torque is increased by using gear transmission to drive the transmission turntable to rotate, so that the corn seeds are conveyed; the conveying mode is rotary conveying;

the image acquisition device consists of an LED annular light source (17), a camera (18) and a connecting screw rod (19); the LED annular light source is arranged on the camera lens and is coaxial with the optical axis of the camera, and the camera is fixed on the dead axle experiment table by the connecting screw rod, so that the optical axis of the camera is vertical to the transmission turntable of the conveying device; the method is characterized in that: the LED annular light source adopts yellow light, the connecting screw is in threaded connection with the camera, and the connecting screw is connected with a fixed shaft of the fixed shaft experiment table through a cross clamp;

the sorting device is divided into two mechanisms, namely a damaged seed removing mechanism [20] and a qualified seed recovering mechanism (21), and specifically comprises a seed dropping groove (22), an opening and closing bottom cover (23), a hinge, a stepping motor c (24), a shifting rod, a connecting screw rod (25) and a magnet; the method is characterized in that: 8 circular holes are respectively formed in a conveying turntable of the conveying device, the included angle between the circle centers of every two adjacent holes is 45 degrees, the seed dropping groove is bonded to the lower sides of the holes of the conveying turntable, two sides of a hinge are respectively connected with the seed dropping groove and an opening and closing bottom cover, and the opening and closing bottom cover is arranged on the lower side of the seed dropping groove through the hinge; 2 magnets are respectively arranged at the opening end of the opening and closing bottom cover and the bottom end groove of the seed dropping groove, the included angle between the magnets and the hinge is 180 degrees, the deflector rod is connected with an output shaft of the stepping motor, and the stepping motor is connected to a fixed shaft of the fixed shaft experiment table through a connecting screw rod and a cross clamp; the horizontal installation included angle between the feeding device and the image acquisition device is 45 degrees, the horizontal installation included angle between the image acquisition device and the damaged seed removing mechanism of the classification device is 90 degrees, the horizontal installation included angle between the damaged seed removing mechanism of the classification device and the uncovering deflector rod of the qualified seed recovery mechanism is 90 degrees, and the horizontal installation angle between the uncovering deflector rod of the qualified seed recovery mechanism and the closing deflector rod is 90 degrees;

the control system is mainly characterized in that: the on-line corn seed damage identification and classification device based on machine vision is controlled by a single chip microcomputer; (1) the lower computer receives the information of the corn seeds and transmits the information to the upper computer, the upper computer sends a signal, and the lower computer controls the image acquisition device to shoot the corn seed image; (2) the lower computer transmits the shot image to the upper computer, and the upper computer judges whether the corn seeds are qualified or not and transmits a judgment result to the lower computer; (3) if the detection result is that the corn seeds are damaged, the lower computer controls the stepping motor c of the damaged seed removing part to drive the shifting rod to rotate by 90 degrees, the opening and closing bottom cover of the damaged seed removing part is opened, and the damaged seeds fall down.

2. The on-line corn seed breakage identifying and classifying device based on the machine vision as claimed in claim 1, wherein the step motor b drives the finger-clip type seed sowing device to perform intermittent motion, the step motor a drives the conveying device to perform intermittent motion, the two intermittent motions are synchronous, and the intermittent time is determined to be in a range of 1-3s according to the detection processing time.

3. The online corn seed breakage identifying and classifying device based on machine vision as claimed in claim 1, wherein the conveying manner is that a stepping motor drives a transmission gear set to perform rotary conveying.

4. The on-line corn seed breakage identifying and classifying device based on the machine vision as claimed in claim 1, wherein the classifying device comprises a breakage seed removing mechanism and a qualified seed recovering mechanism, the removing mechanism is driven by a stepping motor c to rotate a deflector rod by 90 degrees, and an opening and closing bottom cover is opened to enable breakage corn seeds to fall down; the cover opening deflector rod of the qualified seed recovery mechanism opens the opening and closing bottom cover to enable the qualified corn seeds to fall into the recovery seed box below, and the closing deflector rod of the qualified seed recovery mechanism closes the opening and closing bottom cover with the fallen corn seeds to enter the next circle of movement.

5. The online corn seed breakage identifying and sorting device based on machine vision as claimed in claim 1, wherein the control system executes the broken corn seed removing action by controlling the rotation angle of the stepping motor c when receiving the broken seed signal.

Images