CN113670928B - Nondestructive melon and fruit damage identification and hardness detection device based on visual sense of touch - Google Patents
Nondestructive melon and fruit damage identification and hardness detection device based on visual sense of touch Download PDFInfo
- Publication number
- CN113670928B CN113670928B CN202110726957.4A CN202110726957A CN113670928B CN 113670928 B CN113670928 B CN 113670928B CN 202110726957 A CN202110726957 A CN 202110726957A CN 113670928 B CN113670928 B CN 113670928B
- Authority
- CN
- China
- Prior art keywords
- silica gel
- box body
- melon
- fruit
- upper box
- 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
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/01—Arrangements or apparatus for facilitating the optical investigation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/01—Arrangements or apparatus for facilitating the optical investigation
- G01N2021/0106—General arrangement of respective parts
- G01N2021/0112—Apparatus in one mechanical, optical or electronic block
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
- G01N2021/8887—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges based on image processing techniques
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computing Systems (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- Computational Linguistics (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- General Engineering & Computer Science (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- Data Mining & Analysis (AREA)
- Artificial Intelligence (AREA)
- Pathology (AREA)
- Immunology (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- Signal Processing (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The invention relates to a visual sense based nondestructive melon and fruit damage identification and hardness detection device, which is characterized in that: the sampling box comprises a lower box body and an upper box body, wherein the upper box body is arranged above the lower box body in a vertically movable manner, the lower end of the upper box body is provided with an opening, and the upper end of the lower box body is provided with a melon and fruit placing cavity; the silica gel piece is fixed at the opening of the upper box body and comprises a silica gel membrane and a plurality of silica gel synapses, the silica gel membrane covers the opening, and the plurality of silica gel synapses are uniformly distributed on the silica gel membrane at intervals; the camera is fixed with the upper box body and used for collecting the appearance of the melon and fruit to be detected and the relative movement track video of a plurality of silica gel synapses and the upper box body in the contact process of the silica gel piece and the melon and fruit; and the processor is connected with the camera and adopts a deep learning network model to carry out damage identification and hardness detection. The invention can not damage the melon and fruit, has simple structure, low production cost, repeated use and long service life.
Description
Technical Field
The invention relates to the field of nondestructive testing of fruits and vegetables, in particular to a nondestructive melon and fruit damage identification and hardness testing device based on visual sense and touch.
Background
The current stage melon and fruit hardness detection method is divided into destructive detection and nondestructive detection. Destructive testing usually employs a sclerometer probe inserted into the interior of the flesh, however this method can damage the fruit, greatly reducing its storage time. Nondestructive testing usually adopts instruments such as a spectrum instrument or a texture instrument, the testing method is fast, fruits are not damaged, and therefore the nondestructive testing method is widely concerned, but the instruments such as the spectrum instrument and the texture instrument are high in cost and difficult to widely popularize.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provide a visual sense based nondestructive melon and fruit damage identification and hardness detection device which does not damage melons and fruits, has a simple structure, is low in production cost, can be repeatedly used and has a long service life.
In order to solve the technical problems, the technical scheme of the invention is as follows: the utility model provides a harmless melon and fruit damage discernment and hardness testing device based on look sense of touch, its difference lies in:
the sampling box comprises a lower box body and an upper box body, wherein the upper box body is arranged above the lower box body in a vertically movable manner, the lower end of the upper box body is provided with an opening, and the upper end of the lower box body is provided with a melon and fruit placing cavity;
the silica gel piece is fixed at the opening of the upper box body and comprises a silica gel membrane and a plurality of silica gel synapses, the silica gel membrane covers the opening, and the plurality of silica gel synapses are uniformly distributed on the silica gel membrane at intervals;
the camera is fixed with the upper box body and used for collecting the appearance of the melon and fruit to be detected and the relative movement track video of the plurality of silica gel synapses and the upper box body in the contact process of the silica gel pieces and the melon and fruit;
and
the processor is connected with the camera and adopts a deep learning network model combining Resnet18 and LSTM to carry out damage identification and hardness detection;
the sampling box further comprises a driving piece, and the driving piece can drive the upper box body to move upwards so as to open the sampling box; the driving piece can also drive the upper box body to move downwards so as to close the sampling box; the driving piece comprises a base, a stepping motor, a ball screw, a nut, a linear sliding rail and a sliding block, the base vertically extends, the lower box body is fixed at the lower end of the base, the stepping motor is fixed at the upper end of the base, the output shaft of the stepping motor is fixedly connected with the ball screw, the ball screw vertically extends, the surface of the ball screw is in threaded connection with the nut, the linear sliding rail is fixed with the base and is parallel to the ball screw, the sliding block sleeve is arranged on the linear sliding rail, one end of the sliding block is fixed with the nut, the other end of the sliding block is fixed with the upper box body, and the sliding block can drive the upper box body to move up and down the linear sliding rail.
According to the technical scheme, the silica gel membrane is made of soft silica gel.
According to the technical scheme, the hardness of the silica gel membrane is 12A.
According to the technical scheme, the silica gel film is a sheet-shaped transparent silica gel film.
According to the technical scheme, the thickness of the silica gel membrane is 1mm.
According to the technical scheme, the silica gel synapse is fixed on the lower surface of the silica gel membrane.
According to the technical scheme, the silica gel synapse is made of hard silica gel.
According to the technical scheme, the rigidity of the silica gel synapse is 20A.
According to the technical scheme, the silica gel synapse is colored silica gel.
According to the technical scheme, the device also comprises a driving piece which is used for driving the upper box body to move up and down so as to be opened or closed with the lower box body.
Compared with the prior art, the invention has the beneficial characteristics that: this harmless melon and fruit damage discernment and hardness detection device based on look sense of touch, press down to with lower box body closed in-process at last box body, melon and fruit and silica gel piece contact and extrusion silica gel piece, make the silica gel synapse take place to remove, the camera gathers all silica gel synapses and pushes down the removal orbit of in-process at last box body and wait to detect melon and fruit outward appearance video, and give the treater with the video transmission who gathers, the treater adopts degree of depth study network model to damage discernment and hardness detection, can not damage the melon and fruit, moreover, the steam generator is simple in structure, therefore, the production cost is low, and can be repeatedly used, long service life.
Drawings
FIG. 1 is a schematic diagram of an open state of a sampling box of a nondestructive melon and fruit damage identification and hardness detection device based on visual sense of touch according to an embodiment of the invention;
FIG. 2 is a view taken along line A of FIG. 1;
FIG. 3 is a schematic view of a closed state of a sampling box of the nondestructive melon and fruit damage identification and hardness detection device based on visual sense of touch in the embodiment of the present invention, wherein the melon and fruit is not placed in the sampling box;
FIG. 4 is a cross-sectional view B-B of FIG. 3;
FIG. 5 is a schematic view of a closed state of a sampling box of the nondestructive melon and fruit damage identification and hardness detection device based on visual sense of touch, wherein a melon and fruit is placed in the sampling box;
FIG. 6 is a cross-sectional view C-C of FIG. 5;
wherein: the device comprises a lower box body 1, an upper box body 2, a melon and fruit placing plate 3 (301-melon and fruit placing cavity), a 4-silica gel film, a 5-silica gel synapse, a 6-camera, a 7-data transmission line, an 8-fixing frame, a 9-screw, a 10-base, a 11-stepping motor (1101-output shaft), a 12-ball screw, a 13-nut, a 14-linear slide rail, a 15-slider and a 16-kiwi fruit.
Detailed Description
The present invention will be described in further detail with reference to the following detailed description and accompanying drawings.
Referring to fig. 1 to 6, the visual sense based nondestructive melon and fruit damage identification and hardness detection device includes a sample box, a silicone piece, a camera and a processor. The sampling box can open and shut, and the sampling box includes box body 1 and last box body 2 down, and wherein box body 1 is fixed down, goes up box body 2 and can locate box body 1 directly over down with reciprocating. When the upper box body 2 moves upwards, the sampling box is opened gradually; when the upper box body 2 moves downwards, the sampling box is gradually closed. The lower end of the upper box body 2 is opened. The upper end of lower box body 1 is fixed with the melon and fruit and places board 3, and the melon and fruit that is formed with the undercut on placing board 3 places chamber 301. The silica gel piece is fixed in the opening part of the upper box body 2, the silica gel piece comprises a silica gel membrane 4 and a silica gel synapse 5 which are fixed into a whole, the silica gel membrane 4 covers the opening part of the upper box body 2, and the silica gel synapses 5 are arranged on the silica gel membrane 4 in a rectangular array shape. The camera 6 is fixed with the upper box body 2 and is used for collecting relative movement track videos of the melon and fruit appearance to be detected and the silica gel synapse 5 and the upper box body 2 in the contact process of the silica gel piece and the melon and fruit. The processor is connected with the camera 6 through the USB data transmission line 7, and after the data transmission line 7 transmits the video collected by the camera 6 to the processor, the processor adopts a deep learning network model to perform damage identification and hardness detection.
Preferably, in order to make the camera 6 more stable, the camera 6 is fixed in the upper case 2. More preferably, the camera 6 is positioned right above the melon and fruit placing cavity 301 to ensure better shooting effect of the camera 6. To avoid the camera 6 interfering with the silicone piece, the camera 6 is located at the upper end within the upper box 2.
Preferably, the silicone membrane 4 is made of soft silicone to avoid crushing fruits.
Preferably, the silicone synapse 5 is fixed to the lower surface of the silicone membrane 4. More preferably, in order to make the movement of the silicone synapse 5 more obvious during the contact between the silicone part and the melon and fruit, the silicone synapse 5 is made of hard silicone.
Preferably, the silicone membrane 4 has a hardness of 12A, and in order to avoid crushing the melon, the silicone synapse 5 has a hardness of 20A.
Preferably, in order to make the shooting by the camera 6 clearer, the silicone membrane 4 is a sheet-shaped transparent silicone membrane.
Preferably, the silicone membrane 4 has a length of 150mm, a width of 120mm and a thickness of 1mm. The diameter of the silica gel synapse 5 is 2.5mm, the thickness is 3mm, and the distance between two adjacent silica gel synapses 5 is 8mm.
Preferably, in order to make the silicone synapse 5 more clear in the video, the silicone synapse 5 is colored silicone. More preferably, to further highlight the colloidal silica synapse 5, the colloidal silica synapse 5 is black.
Specifically, the sampling box still includes the fixed frame 8 that is used for fixing pellosil 4 on last box body 2, and pellosil 4 is pressed from both sides between fixed frame 8 and last box body 2, and fixed frame 8 passes through screw 9 with last box body 2 and fixes.
Preferably, in order to stably move the upper box body 2, the device further comprises a driving part which can drive the upper box body 2 to move upwards so as to open the sampling box; the driving piece can also drive the upper box body 2 to move downwards, so that the sampling box is closed. In some embodiments of the present invention, the driving member includes a base 10, a stepping motor 11, a ball screw 12, a nut 13, a linear slide 14, and a slider 15, the base 10 extends vertically, the lower case 1 is fixed at the lower end of the base 10, the stepping motor 11 is fixed at the upper end of the base 10, an output shaft 1101 of the stepping motor 11 is fixedly connected with the ball screw 12, the ball screw 12 extends vertically, the nut 13 is connected to a surface of the ball screw 12 in a threaded manner, the linear slide 14 is fixed with the base 10 and is arranged parallel to the ball screw 12, the slider 15 is sleeved on the linear slide 14, one end of the slider 15 is fixed with the nut 13, the other end of the slider 15 is fixed with the upper case 2, and under the action of the stepping motor 11, the slider 15 can drive the upper case 2 to move up and down along the linear slide 14. More preferably, the linear slides 14 are two and symmetrically disposed about the ball screw 12, and the slide 15 is one and respectively engaged with the two linear slides 14.
Referring to fig. 1 to 6, taking kiwi fruit 16 as an example, the nondestructive melon and fruit damage identification and hardness detection device based on visual sense of touch according to the embodiment of the present invention has the following steps: please refer to fig. 1, open the sampling box, put kiwi fruit 16 into melon and fruit placement cavity 301, make the upper half surface of kiwi fruit 16 just face camera 6 and place, then close the sampling box, please refer to fig. 5 and fig. 6, in the closed process of sampling box, camera 6 gathers kiwi fruit 16 outward appearance, and the relative movement orbit video of silica gel synapse 5 and last box body 2 in the contact process of silica gel spare and kiwi fruit 16 and transmits for the treater, the sampling is accomplished, the treater adopts the degree of deep learning network model to carry out damage identification and hardness detection. Gather the video and accomplish the back, upset kiwi fruit 16 from top to bottom, the latter half upset that makes kiwi fruit 16 is upwards placed towards camera 6 promptly, then closed sampling box, at the closed in-process of sampling box, 16 outward appearances of kiwi fruit are gathered to camera 6, and the relative movement orbit video of 16 contact in-process silica gel synapses 5 of silica gel spare and last box body 2 of kiwi fruit transmits for the treater, the sampling is accomplished, the treater adopts degree of deep learning network model to damage discernment and hardness detection. The hardness of the kiwi fruit 16 is an average value of the results obtained by two hardness tests.
In a preferred embodiment of the invention, the deep learning model is a Resnet18 and LSTM combined deep learning model based on sequence multi-imaging, which is used for evaluating hardness and damage or lesion identification of melons and fruits. The concrete contents and steps of the model are as follows:
a) An image acquisition stage: the camera 6 is connected with the processor through a USB data transmission line 7, collects melon and fruit and silica gel synapse contact videos, and meanwhile, an Opencv library is called by a Python development environment operated by the processor, receives the videos recorded by the camera 6, and performs frame extraction on the videos so as to perform LSTM frame extraction processing and feature extraction on the videos.
b) A pretreatment stage: the picture resolution obtained by LSTM frame extraction processing of the video is uniformly processed into 224 × 224 pixels in a Python development environment, and normalization processing is carried out.
c) A characteristic extraction stage: the first frame, the intermediate frame and the last frame of the video extraction are respectively input into a Resnet18 model which is trained in advance, the video frames continuously carry out various convolution layer, pooling layer and full connection layer feature extraction operations in the model, and 1000 image features are extracted from the last full connection layer.
d) And a hardness prediction stage: and (3) sending the obtained picture extraction features of each frame of the video to an LSTM model, wherein each LSTM layer outputs 128 features, and all the LSTM layers are interconnected in sequence. That is, the result of the LSTM operation on the previous frame image is passed to the LSTM operation on the next frame until the last frame, so that the features of all video frames can be utilized. Finally, the kiwi fruit hardness can be predicted after the characteristics pass through a full-connection layer.
e) And (3) lesion and injury identification stage: the Python development environment processes the last frame of picture of the video frame to 224 × 224 resolution and inputs the picture into another Resnet18 model to judge whether the melon and fruit has lesion or damage.
The foregoing is a detailed description of the present invention with reference to specific embodiments, and it should not be considered that the present invention is limited to the specific embodiments, and it will be understood by those skilled in the art that various changes and modifications may be made without departing from the spirit and scope of the present invention.
Claims (9)
1. The utility model provides a harmless melon and fruit damage discernment and hardness detection device based on look sense of touch which characterized in that:
the sampling box comprises a lower box body and an upper box body, wherein the upper box body is arranged above the lower box body in a vertically movable manner, the lower end of the upper box body is provided with an opening, and the upper end of the lower box body is provided with a melon and fruit placing cavity;
the silica gel piece is fixed at the opening of the upper box body and comprises a silica gel membrane and a plurality of silica gel synapses, the silica gel membrane covers the opening, and the plurality of silica gel synapses are uniformly distributed on the silica gel membrane at intervals;
the camera is fixed with the upper box body and used for collecting the appearance of the melon and fruit to be detected and the relative movement track video of the plurality of silica gel synapses and the upper box body in the contact process of the silica gel pieces and the melon and fruit;
and
the processor is connected with the camera and adopts a deep learning network model combining Resnet18 and LSTM to carry out damage identification and hardness detection;
the sampling box further comprises a driving piece, and the driving piece can drive the upper box body to move upwards so as to open the sampling box; the driving piece can also drive the upper box body to move downwards so as to close the sampling box; the driving piece comprises a base, a stepping motor, a ball screw, a nut, a linear sliding rail and a sliding block, wherein the base vertically extends, the lower box body is fixed at the lower end of the base, the stepping motor is fixed at the upper end of the base, the output shaft of the stepping motor is fixedly connected with the ball screw, the ball screw vertically extends, the surface of the ball screw is in threaded connection with the nut, the linear sliding rail is fixed with the base and is arranged in parallel with the ball screw, the sliding block sleeve is arranged on the linear sliding rail, one end of the sliding block is fixed with the nut, the other end of the sliding block is fixed with the upper box body, and the sliding block can drive the upper box body to move along the linear sliding rail up and down under the action of the stepping motor.
2. An optotactility-based nondestructive melon and fruit damage identification and hardness detection device as claimed in claim 1, wherein: the silica gel membrane is made of soft silica gel.
3. An optotactual-based nondestructive melon and fruit damage identification and hardness detection device as claimed in claim 2, wherein: the hardness of the silica gel membrane is 12A.
4. An optotactual-based nondestructive melon and fruit damage identification and hardness detection device as claimed in claim 1, wherein: the silica gel membrane is a flaky transparent silica gel membrane.
5. An optotactual-based nondestructive melon and fruit damage identification and hardness detection device as claimed in claim 1, wherein: the thickness of pellosil is 1mm.
6. An optotactual-based nondestructive melon and fruit damage identification and hardness detection device as claimed in claim 1, wherein: the silica gel synapse is fixed on the lower surface of the silica gel membrane.
7. An optotactual-based nondestructive melon and fruit damage identification and hardness detection device as claimed in claim 6, wherein: the silica gel synapse is made of hard silica gel.
8. An optotactual-based nondestructive melon and fruit damage identification and hardness detection device as claimed in claim 7, wherein: the silica gel synapse has a hardness of 20A.
9. An optotactual-based nondestructive melon and fruit damage identification and hardness detection device as claimed in claim 1, wherein: the silica gel synapse is colored silica gel.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110726957.4A CN113670928B (en) | 2021-06-29 | 2021-06-29 | Nondestructive melon and fruit damage identification and hardness detection device based on visual sense of touch |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110726957.4A CN113670928B (en) | 2021-06-29 | 2021-06-29 | Nondestructive melon and fruit damage identification and hardness detection device based on visual sense of touch |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113670928A CN113670928A (en) | 2021-11-19 |
| CN113670928B true CN113670928B (en) | 2023-03-28 |
Family
ID=78538365
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110726957.4A Active CN113670928B (en) | 2021-06-29 | 2021-06-29 | Nondestructive melon and fruit damage identification and hardness detection device based on visual sense of touch |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113670928B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117621133B (en) * | 2023-12-29 | 2024-10-18 | 求之科技(无锡)有限公司 | Multi-finger visual touch track clamping jaw and control method thereof |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109744785B (en) * | 2019-02-22 | 2023-11-24 | 浙江大学 | Intelligent self-adaptive mattress based on flexible pressure sensor array |
| CN110162175B (en) * | 2019-05-16 | 2022-04-19 | 腾讯科技（深圳）有限公司 | Vision-based tactile measurement method, device, equipment and storage medium |
| CN111150373B (en) * | 2020-02-17 | 2020-11-24 | 中国科学院力学研究所 | Device and method for measuring hardness of muscle or soft tissue |
| CN213209793U (en) * | 2020-09-30 | 2021-05-14 | 安徽汉碟电子材料有限公司 | Heat conduction silica gel piece hardness detection device |
| CN112485140B (en) * | 2020-11-06 | 2022-02-18 | 浙江大学 | Fruit hardness sensor integrated on flexible finger |
| CN113834727B (en) * | 2021-09-02 | 2024-04-12 | 浙江大学 | Fruit maturity tactile sensing device and detection method thereof |
-
2021
- 2021-06-29 CN CN202110726957.4A patent/CN113670928B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN113670928A (en) | 2021-11-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Luo et al. | Temporal and spatial deep learning network for infrared thermal defect detection | |
| CN112858473B (en) | Turnout switch blade damage state monitoring method based on feature fusion | |
| CN103439270B (en) | A kind of high spectrum image detects the method that Peach fruits damages to plants caused by sudden drop in temperature in early days | |
| Sharkawy et al. | Particle swarm optimization feature selection for the classification of conducting particles in transformer oil | |
| CN109460618A (en) | A kind of rolling bearing remaining life on-line prediction method and system | |
| CN113670928B (en) | Nondestructive melon and fruit damage identification and hardness detection device based on visual sense of touch | |
| Wang et al. | Non-negative EMD manifold for feature extraction in machinery fault diagnosis | |
| CN112990082B (en) | Detection and identification method of underwater sound pulse signal | |
| Wu et al. | An end-to-end learning method for industrial defect detection | |
| Mumtaz et al. | Computer aided visual inspection of aircraft surfaces | |
| Wang et al. | Non-destructive detection of apple multi-quality parameters based on hyperspectral imaging technology and 3D-CNN. | |
| Dimiccoli et al. | Recognizing object surface material from impact sounds for robot manipulation | |
| CN107607624A (en) | It is adapted to the acoustic emission detection system and method for high frequency motion surface of friction pair | |
| Li et al. | Assessing fruit hardness in robot hands using electric gripper actuators with tactile sensors | |
| Liu et al. | Bearing performance degradation assessment using linear discriminant analysis and coupled HMM | |
| Qiu et al. | A novel method for detection of internal quality of walnut kernels using low-field magnetic resonance imaging | |
| Jayanthi et al. | Fruit quality inspection system using image processing | |
| CN115078538B (en) | High-voltage cable aluminum sheath corrosion ultrasonic guided wave detection method, system and medium based on SWT-AE | |
| Li et al. | A robust fault diagnosis method for rolling bearings based on deep convolutional neural network | |
| Marin et al. | Image-based Fruit Recognition and Classification | |
| Li et al. | A novel tomato volume measurement method based on machine vision | |
| Zhang et al. | On-line sorting maturity of cherry tomato bymachine vision | |
| Sun et al. | Research on intelligent fault diagnosis method for rolling bearings based on multi-feature fusion of CWT-2D-CNN | |
| CN109199432B (en) | Parallel compression sensing method for multi-path synchronous acquisition of heart sound signals | |
| CN109001244B (en) | Device and method for detecting potato quality based on infrared thermal imaging technology |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |