CN109060806B - Chest bottom end material type recognition mechanism - Google Patents
Chest bottom end material type recognition mechanism Download PDFInfo
- Publication number
- CN109060806B CN109060806B CN201810991967.9A CN201810991967A CN109060806B CN 109060806 B CN109060806 B CN 109060806B CN 201810991967 A CN201810991967 A CN 201810991967A CN 109060806 B CN109060806 B CN 109060806B
- Authority
- CN
- China
- Prior art keywords
- image
- bottom end
- chest
- data processing
- frequency domain
- 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
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
-
- 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
- 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
Abstract
The present invention relates to a kind of chest bottom end material type recognition mechanisms, comprising: the bottom end of chest is arranged in light quantity detection device, and the light quantity for the bottom position to chest detects, and to obtain corresponding bottom end light quantity numerical value, and exports the bottom end light quantity numerical value；LED illumination device is connect with the light quantity detection device, for when the bottom end light quantity numerical value is less than or equal to default light quantity threshold value, automatic luminous to be to realize the lighting operation to the bottom position of chest；The bottom end of chest is arranged in high-definition shooting equipment, and positioned at the side of the LED illumination device, the material of the lower section for the bottom end to chest carries out high-definition shooting operation, to obtain and export corresponding high definition images of materials；Big data server, for carrying out the identification processing based on images of materials feature to image, to obtain the corresponding material type of image, using the material type of the lower section of the bottom end as chest.By means of the invention it is possible to which effective supply chest pushes power-assisted.
Description
Technical field
The present invention relates to big data processing field more particularly to a kind of chest bottom end material type recognition mechanisms.
Background technique
The data mining of big data unlike the statistics and analysis process of front, data mining it is typically no what in advance
The theme set carries out the calculating based on various algorithms mainly on available data, to play prediction (Predict)
Effect, to realize the demand of some high-level data analyses.
Comparing typical algorithm has the Kmeans for cluster, the SVM for statistical learning and for classification
NaiveBayes, main tool to be used have the Mahout etc. of Hadoop.The characteristics of process and challenge are mainly used for excavating
Algorithm it is very complicated, and calculate that the data volume that is related to and calculation amount are all very big, and frequently-used data mining algorithm is all with single thread
It is main.
Summary of the invention
In order to solve the difficult technical problem of current chest push, the present invention provides a kind of chest bottom end material types to distinguish
Know mechanism, on the basis of big data processing, is accurately determined by targeted image procossing mechanism and image recognition mechanism
The material type of the lower section of the bottom end of chest, and the material type of the lower section of the bottom end based on chest further determines that corresponding help
Power size, to realize that the adaptive power-assisted to chest acts；Resolution based on image to be processed is in the image to be processed
The center image block of the image to be processed is set, determines the quantity of the image-region in the image to be processed in
Number of objects is entreated, the ratio of whole number of objects in the image to be processed is occupied based on the central object quantity, is formulated not
Same image filtering strategy, improves the cost performance of image real time transfer.
According to an aspect of the present invention, a kind of chest bottom end material type recognition mechanism is provided, the mechanism includes:
The bottom end of chest is arranged in light quantity detection device, and the light quantity for the bottom position to chest detects, to obtain
Corresponding bottom end light quantity numerical value is obtained, and exports the bottom end light quantity numerical value；LED illumination device connects with the light quantity detection device
It connects, for when the bottom end light quantity numerical value is less than or equal to default light quantity threshold value, automatic luminous to be to realize the bottom end position to chest
The lighting operation set；The bottom end of chest is arranged in high-definition shooting equipment, positioned at the side of the LED illumination device, for case
The material of the lower section of the bottom end of son carries out high-definition shooting operation, to obtain and export corresponding high definition images of materials.
More specifically, in the chest bottom end material type recognition mechanism, further includes:
The side of the high-definition shooting equipment is arranged in homomorphic filtering equipment, for receiving the high definition images of materials, and
Homomorphic filtering processing is executed to the high definition images of materials, to obtain homomorphic filtering image corresponding with the image to be processed,
And export the homomorphic filtering image；Processing executes equipment, connect with the homomorphic filtering equipment, for receiving the homomorphism filter
Wave image executes smoothing processing to the homomorphic filtering image, to obtain smoothing processing corresponding with the homomorphic filtering image
Image, and export the smoothing processing image.
More specifically, in the chest bottom end material type recognition mechanism, further includes:
First data processing equipment executes equipment with the processing and connect, and for receiving the smoothing processing image, obtains
Each edge pixel point in the smoothing processing image determines the smoothing processing image based on each edge pixel point
In one or more objects respectively where one or more image-regions；Second data processing equipment, with first number
It is connected according to processing equipment, for receiving the quantity of one or more of image-regions using as region quantity, and based on described
The resolution of smoothing processing image sets the center image block of the smoothing processing image in the smoothing processing image, determines
The quantity of image-region in the center image block as central object quantity to export；Third data processing equipment, with
The second data processing equipment connection, for receiving the central object quantity and the region quantity, calculates the center
Number of objects occupies the ratio of the region quantity, and the divergence based on the corresponding Gaussian curve of the ratio-dependent；4th
Data processing equipment is connect with the third data processing equipment, for selecting each of frequency domain frequency domain to be processed point
Filter factor is as follows: determine the frequency domain point to be processed to frequency domain origin distance, obtain the square value of the distance using as
First square value determines the square value of the divergence for the Gaussian curve that the third data processing equipment determines, by the Gauss
The square value of the divergence of curve multiplied by 2 using as the second square value, by first square value divided by second square value
Result afterwards carries out taking negative operation to obtain exponential depth, and executing the exponential depth using natural logrithm is the index budget at bottom to obtain
Obtain the filter factor of the frequency domain point to be processed；5th data processing equipment, respectively with first data processing equipment and institute
The connection of the 4th data processing equipment is stated, for executing the smoothing processing image from Fourier transformation, to obtain corresponding frequency
Rate domain signal determines the frequency domain value of each of frequency domain signal frequency domain point multiplied by the 4th data processing equipment
The frequency domain point filter factor to obtain the filter in frequency domain value of the frequency domain point；6th data processing equipment, with described
The connection of five data processing equipments, to obtain each filter in frequency domain value of each frequency domain point in the frequency domain signal, and is based on
Each filter in frequency domain value of each frequency domain point in the frequency domain signal carry out frequency domain obtained to the transformation of time-domain and
The corresponding data processing image of the smoothing processing image；Big data server is connect with the 6th data processing equipment, is used
In receiving the data processing image, the processing of the identification based on images of materials feature is carried out to the data processing image, to obtain
The corresponding material type of the data processing image is obtained, using the material type of the lower section of the bottom end as chest；Power-assisted driving is set
It is standby, it is connect with the big data server, the material type of the lower section of the bottom end for receiving chest, and the bottom end based on chest
The material type of lower section determine corresponding power-assisted size；Wherein, in the power-assisted driving equipment, the bottom end based on chest
The material type of lower section determines that corresponding power-assisted size includes: that the material type bring resistance of the lower section of the bottom end of chest is got over
Greatly, the corresponding power-assisted determined is bigger.
More specifically, in the chest bottom end material type recognition mechanism, further includes:
SD storage card is connect, for depositing respectively with first data processing equipment and the third data processing equipment
Store up the default deviation threshold value.
More specifically, in the chest bottom end material type recognition mechanism: first data processing equipment obtains institute
Stating each edge pixel point in smoothing processing image includes: that red color component is deviateed each pixel red color of surrounding
Component mean value is determined as edge pixel point up to the default pixel for deviateing threshold value.
More specifically, in the chest bottom end material type recognition mechanism: second data processing equipment is based on institute
The resolution for stating smoothing processing image sets the center image block packet of the smoothing processing image in the smoothing processing image
Include: the resolution of the smoothing processing image is lower, and the center image block of the smoothing processing image of setting is smaller.
More specifically, in the chest bottom end material type recognition mechanism: the SD storage card is also used to store described
Central object quantity occupies the corresponding relationship of the ratio of the region quantity and the divergence of Gaussian curve.
More specifically, in the chest bottom end material type recognition mechanism: the LED illumination device is also used to described
When bottom end light quantity numerical value is greater than the default light quantity threshold value, automatic black out is to stop the lighting operation to the bottom position of chest.
Detailed description of the invention
Embodiment of the present invention is described below with reference to attached drawing, in which:
Fig. 1 is the knot of the chest according to applied by the chest bottom end material type recognition mechanism shown in embodiment of the present invention
Structure schematic diagram.
Specific embodiment
The embodiment of chest bottom end material type recognition mechanism of the invention is carried out specifically below with reference to accompanying drawings
It is bright.
The statistics of big data processing and analysis are mainly using distributed data base or distributed computing cluster come to storage
Mass data in the inner carries out common analysis and Classifying Sum etc., to meet most of common analysis demands, in this side
Face, some real-time demands can use the Exadata of GreenPlum, Oracle of EMC, and the storage of the column based on MySQL
Infobright etc., and Hadoop can be used in some batch processings, or the demand based on semi-structured data.
Counting with the main feature and challenge for analyzing this part is that the data volume that analysis is related to is big, special to system resource
It is not that I/O has great occupancy.In order to overcome above-mentioned deficiency, the present invention has built a kind of chest bottom end material type identifier
Structure can effectively solve the problem that corresponding technical problem.
Fig. 1 is the knot of the chest according to applied by the chest bottom end material type recognition mechanism shown in embodiment of the present invention
Structure schematic diagram.
The chest bottom end material type recognition mechanism shown according to an embodiment of the present invention includes:
The bottom end of chest is arranged in light quantity detection device, and the light quantity for the bottom position to chest detects, to obtain
Corresponding bottom end light quantity numerical value is obtained, and exports the bottom end light quantity numerical value；
LED illumination device is connect with the light quantity detection device, default for being less than or equal in the bottom end light quantity numerical value
When light quantity threshold value, automatic luminous is to realize the lighting operation to the bottom position of chest；
The bottom end of chest is arranged in high-definition shooting equipment, positioned at the side of the LED illumination device, for chest
The material of the lower section of bottom end carries out high-definition shooting operation, to obtain and export corresponding high definition images of materials.
Then, continue to carry out further the specific structure of chest bottom end material type recognition mechanism of the invention
It is bright.
In the chest bottom end material type recognition mechanism, further includes:
The side of the high-definition shooting equipment is arranged in homomorphic filtering equipment, for receiving the high definition images of materials, and
Homomorphic filtering processing is executed to the high definition images of materials, to obtain homomorphic filtering image corresponding with the image to be processed,
And export the homomorphic filtering image；
Processing executes equipment, connect with the homomorphic filtering equipment, for receiving the homomorphic filtering image, to described same
State filtering image executes smoothing processing, to obtain smoothing processing image corresponding with the homomorphic filtering image, and described in output
Smoothing processing image.
In the chest bottom end material type recognition mechanism, further includes:
First data processing equipment executes equipment with the processing and connect, and for receiving the smoothing processing image, obtains
Each edge pixel point in the smoothing processing image determines the smoothing processing image based on each edge pixel point
In one or more objects respectively where one or more image-regions；
Second data processing equipment is connect, for receiving one or more of figures with first data processing equipment
As region quantity using the resolution as region quantity, and based on the smoothing processing image in the smoothing processing image
Set the center image block of the smoothing processing image, determine the quantity of the image-region in the center image block using as
The output of central object quantity；
Third data processing equipment is connect with second data processing equipment, for receiving the central object quantity
With the region quantity, the ratio that the central object quantity occupies the region quantity is calculated, and is based on the ratio-dependent
The divergence of corresponding Gaussian curve；
4th data processing equipment is connect with the third data processing equipment, for each of frequency domain wait locate
Reason frequency domain point selection filter factor is as follows: determining that the frequency domain point to be processed to the distance of frequency domain origin, obtains the distance
Square value is to determine square of the divergence for the Gaussian curve that the third data processing equipment determines as the first square value
Value, by the square value of the divergence of the Gaussian curve multiplied by 2 using as the second square value, by first square value divided by institute
Result after stating the second square value carries out taking negative operation to obtain exponential depth, executes the exponential depth using natural logrithm the bottom of as
Index budget is to obtain the filter factor of the frequency domain point to be processed；
5th data processing equipment connects with first data processing equipment and the 4th data processing equipment respectively
It connects, for executing the smoothing processing image from Fourier transformation, to obtain corresponding frequency domain signal, to the frequency domain
The filtering system for the frequency domain point that the frequency domain value of each of signal frequency domain point is determined multiplied by the 4th data processing equipment
It counts to obtain the filter in frequency domain value of the frequency domain point；
6th data processing equipment is connect with the 5th data processing equipment, to obtain in the frequency domain signal
Each filter in frequency domain value of each frequency domain point, and each filter in frequency domain value based on each frequency domain point in the frequency domain signal
It carries out frequency domain and obtains data processing image corresponding with the smoothing processing image to the transformation of time-domain；
Big data server is connect with the 6th data processing equipment, for receiving the data processing image, to institute
It states data processing image and carries out the identification processing based on images of materials feature, to obtain the corresponding material of the data processing image
Type, using the material type of the lower section of the bottom end as chest；
Power-assisted driving equipment is connect with the big data server, the material class of the lower section of the bottom end for receiving chest
Type, and corresponding power-assisted size is determined based on the material type of the lower section of the bottom end of chest；
Wherein, in the power-assisted driving equipment, corresponding help is determined based on the material type of the lower section of the bottom end of chest
Power size includes: that the material type bring resistance of the lower section of the bottom end of chest is bigger, and the corresponding power-assisted determined is bigger.
In the chest bottom end material type recognition mechanism, further includes:
SD storage card is connect, for depositing respectively with first data processing equipment and the third data processing equipment
Store up the default deviation threshold value.
In the chest bottom end material type recognition mechanism: first data processing equipment obtains the smoothing processing
Each edge pixel point in image includes: that red color component is deviateed each pixel red color component mean value of surrounding to reach
The default pixel for deviateing threshold value is determined as edge pixel point.
In the chest bottom end material type recognition mechanism: second data processing equipment is based on the smoothing processing
The center image block that the resolution of image sets the smoothing processing image in the smoothing processing image includes: described smooth
The resolution for handling image is lower, and the center image block of the smoothing processing image of setting is smaller.
In the chest bottom end material type recognition mechanism: the SD storage card is also used to store the central object number
Amount occupies the corresponding relationship of the ratio of the region quantity and the divergence of Gaussian curve.
In the chest bottom end material type recognition mechanism: the LED illumination device is also used in the bottom end light quantity
When numerical value is greater than the default light quantity threshold value, automatic black out is to stop the lighting operation to the bottom position of chest.
In addition, DDR memory device can be used and replace the SD storage in the chest bottom end material type recognition mechanism
Card.
DDR=Double Data Rate Double Data Rate synchronous DRAM.Strictly speaking DDR should be DDR
SDRAM, people's habit are known as DDR, wherein SDRAM is Synchronous Dynamic Random Access Memory
Abbreviation, i.e. Synchronous Dynamic Random Access Memory.And DDR SDRAM is the abbreviation of Double Data Rate SDRAM, is double
The meaning of times rate synchronous DRAM.DDR memory is developed on the basis of sdram memory, is still continued to use
SDRAM production system, therefore for memory manufacturer, only the equipment for manufacturing common SDRAM need to slightly be improved, be can be realized
The production of DDR memory can effectively reduce cost.
Double Data Rate: compared with traditional single data rate, DDR technology realize in a clock cycle into
Row read/write operation twice, i.e., execute a read/write operation in the rising edge of clock and failing edge respectively.
Using chest bottom end material type recognition mechanism of the invention, time-consuming expense is manually pushed for chest in the prior art
It is the technical issues of power, quasi- by targeted image procossing mechanism and image recognition mechanism on the basis of big data processing
Determine the material type of the lower section of the bottom end of chest, and the material type of the lower section of the bottom end based on chest further determines that pair
The power-assisted size answered, to realize that the adaptive power-assisted to chest acts；Resolution based on image to be processed is described to be processed
Set the center image block of the image to be processed in image, determine the quantity of the image-region in the image to be processed with
As central object quantity, the ratio of whole number of objects in the image to be processed is occupied based on the central object quantity,
Different image filtering strategies is formulated, the cost performance of image real time transfer is improved, to solve above-mentioned technical problem.
It is understood that although the present invention has been disclosed in the preferred embodiments as above, above-described embodiment not to
Limit the present invention.For any person skilled in the art, without departing from the scope of the technical proposal of the invention,
Many possible changes and modifications all are made to technical solution of the present invention using the technology contents of the disclosure above, or are revised as
With the equivalent embodiment of variation.Therefore, anything that does not depart from the technical scheme of the invention are right according to the technical essence of the invention
Any simple modifications, equivalents, and modifications made for any of the above embodiments still fall within the range of technical solution of the present invention protection
It is interior.
Claims (2)
1. a kind of chest bottom end material type recognition mechanism, which is characterized in that the mechanism includes:
The bottom end of chest is arranged in light quantity detection device, and the light quantity for the bottom position to chest detects, with acquisition pair
The bottom end light quantity numerical value answered, and export the bottom end light quantity numerical value；
LED illumination device is connect with the light quantity detection device, for being less than or equal to default light quantity in the bottom end light quantity numerical value
When threshold value, automatic luminous is to realize the lighting operation to the bottom position of chest；
The bottom end of chest is arranged in high-definition shooting equipment, positioned at the side of the LED illumination device, for the bottom end to chest
Lower section material carry out high-definition shooting operation, to obtain and export corresponding high definition images of materials；
The side of the high-definition shooting equipment is arranged in homomorphic filtering equipment, for receiving the high definition images of materials, and to institute
It states high definition images of materials and executes homomorphic filtering processing, to obtain homomorphic filtering image corresponding with image to be processed, and export institute
State homomorphic filtering image；
Processing executes equipment, connect with the homomorphic filtering equipment, for receiving the homomorphic filtering image, filters to the homomorphism
Wave image executes smoothing processing, to obtain smoothing processing image corresponding with the homomorphic filtering image, and exports described smooth
Handle image；
First data processing equipment executes equipment with the processing and connect, for receiving the smoothing processing image, described in acquisition
Each edge pixel point in smoothing processing image is determined in the smoothing processing image based on each edge pixel point
One or more image-regions where one or more objects difference；
Second data processing equipment is connect, for receiving one or more of image districts with first data processing equipment
The quantity in domain is set in the smoothing processing image using the resolution as region quantity, and based on the smoothing processing image
The center image block of the smoothing processing image determines the quantity of image-region in the center image block using as center
Number of objects output；
Third data processing equipment is connect with second data processing equipment, for receiving the central object quantity and institute
Region quantity is stated, calculates the ratio that the central object quantity occupies the region quantity, and corresponding based on the ratio-dependent
Gaussian curve divergence；
4th data processing equipment is connect with the third data processing equipment, for each of frequency domain frequency to be processed
Point selection filter factor in domain is as follows: determining that the frequency domain point to be processed to the distance of frequency domain origin, obtains square of the distance
Value, will to determine the square value of the divergence for the Gaussian curve that the third data processing equipment determines as the first square value
The square value of the divergence of the Gaussian curve multiplied by 2 using as the second square value, by first square value divided by described
Result after two square values carries out taking negative operation to obtain exponential depth, executes the exponential depth using natural logrithm as the index at bottom
Budget is to obtain the filter factor of the frequency domain point to be processed；
5th data processing equipment is connect with first data processing equipment and the 4th data processing equipment respectively, is used
It executes in the smoothing processing image from Fourier transformation, to obtain corresponding frequency domain signal, to the frequency domain signal
Each of frequency domain point frequency domain value multiplied by the 4th data processing equipment determine the frequency domain point filter factor with
Obtain the filter in frequency domain value of the frequency domain point；
6th data processing equipment is connect with the 5th data processing equipment, each in the frequency domain signal to obtain
Each filter in frequency domain value of frequency domain point, and carried out based on each filter in frequency domain value of each frequency domain point in the frequency domain signal
Frequency domain obtains data processing image corresponding with the smoothing processing image to the transformation of time-domain；
Big data server is connect with the 6th data processing equipment, for receiving the data processing image, to the number
The processing of the identification based on images of materials feature is carried out according to processing image, to obtain the corresponding material class of the data processing image
Type, using the material type of the lower section of the bottom end as chest；
Power-assisted driving equipment is connect with the big data server, the material type of the lower section of the bottom end for receiving chest, and
Corresponding power-assisted size is determined based on the material type of the lower section of the bottom end of chest；
Wherein, in the power-assisted driving equipment, determine that corresponding power-assisted is big based on the material type of the lower section of the bottom end of chest
It is small include: the lower section of the bottom end of chest material type bring resistance it is bigger, determine corresponding power-assisted it is bigger.
2. chest bottom end material type recognition mechanism as described in claim 1, which is characterized in that the mechanism further include:
SD storage card is connect, for storing respectively with first data processing equipment and the third data processing equipment
State default deviation threshold value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810991967.9A CN109060806B (en) | 2018-08-29 | 2018-08-29 | Chest bottom end material type recognition mechanism |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810991967.9A CN109060806B (en) | 2018-08-29 | 2018-08-29 | Chest bottom end material type recognition mechanism |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109060806A CN109060806A (en) | 2018-12-21 |
| CN109060806B true CN109060806B (en) | 2019-09-13 |
Family
ID=64757620
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810991967.9A Active CN109060806B (en) | 2018-08-29 | 2018-08-29 | Chest bottom end material type recognition mechanism |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109060806B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109714534A (en) * | 2019-01-02 | 2019-05-03 | 赵欣 | Medical device data mechanism for monitoring |
| CN110308868B (en) * | 2019-03-19 | 2020-01-31 | 武汉烽火信息集成技术有限公司 | Self-adaptive big data storage platform |
| CN111240583B (en) * | 2019-03-19 | 2020-12-08 | 陈俊龙 | Self-adaptive big data storage method |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101216435A (en) * | 2008-01-03 | 2008-07-09 | 东华大学 | Fabric flaw automatic detection method based on multi-fractal characteristic parameter |
| CN103842800A (en) * | 2011-09-30 | 2014-06-04 | 3M创新有限公司 | Web inspection calibration system and related methods |
| CN105850114A (en) * | 2013-12-27 | 2016-08-10 | 汤姆逊许可公司 | Method for inverse tone mapping of an image |
| CN108021921A (en) * | 2017-11-23 | 2018-05-11 | 塔普翊海（上海）智能科技有限公司 | Image characteristic point extraction system and its application |
| CN108286938A (en) * | 2018-02-06 | 2018-07-17 | 陕西理工大学 | A kind of Machine Vision Inspecting System |
-
2018
- 2018-08-29 CN CN201810991967.9A patent/CN109060806B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101216435A (en) * | 2008-01-03 | 2008-07-09 | 东华大学 | Fabric flaw automatic detection method based on multi-fractal characteristic parameter |
| CN103842800A (en) * | 2011-09-30 | 2014-06-04 | 3M创新有限公司 | Web inspection calibration system and related methods |
| CN105850114A (en) * | 2013-12-27 | 2016-08-10 | 汤姆逊许可公司 | Method for inverse tone mapping of an image |
| CN108021921A (en) * | 2017-11-23 | 2018-05-11 | 塔普翊海（上海）智能科技有限公司 | Image characteristic point extraction system and its application |
| CN108286938A (en) * | 2018-02-06 | 2018-07-17 | 陕西理工大学 | A kind of Machine Vision Inspecting System |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109060806A (en) | 2018-12-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109060806B (en) | Chest bottom end material type recognition mechanism | |
| TWI746674B (en) | Type prediction method, device and electronic equipment for identifying objects in images | |
| CN109670528B (en) | Data expansion method facing pedestrian re-identification task and based on paired sample random occlusion strategy | |
| Mathias et al. | Handling occlusions with franken-classifiers | |
| CN111860494B (en) | Optimization method and device for image target detection, electronic equipment and storage medium | |
| WO2021238455A1 (en) | Data processing method and device, and computer-readable storage medium | |
| Sinha et al. | Class-wise difficulty-balanced loss for solving class-imbalance | |
| WO2021114916A1 (en) | Risk detection method, apparatus and device | |
| CN110197208A (en) | A kind of textile flaw intelligent measurement classification method and device | |
| CN109086753A (en) | Traffic sign recognition method, device based on binary channels convolutional neural networks | |
| Muttenthaler et al. | Human alignment of neural network representations | |
| CN103679192A (en) | Image scene type discrimination method based on covariance features | |
| CN110826618A (en) | Personal credit risk assessment method based on random forest | |
| CN102938151A (en) | Method for detecting anomaly of hyperspectral image | |
| CN109635643A (en) | A kind of fast human face recognition based on deep learning | |
| CN110163631A (en) | The accurate method of servicing of client, device, computer equipment and storage medium | |
| CN108154163A (en) | Data processing method, data identification and learning method and its device | |
| CN106339665A (en) | Fast face detection method | |
| CN111178435A (en) | Classification model training method and system, electronic equipment and storage medium | |
| CN109709093A (en) | Chest bottom end material type discrimination method | |
| CN103324923A (en) | Handwritten character recognition method based on sparse representation | |
| Xu et al. | Bidirectional matrix feature pyramid network for object detection | |
| US20230229897A1 (en) | Distances between distributions for the belonging-to-the-distribution measurement of the image | |
| Choi et al. | Utilizing class information for deep network representation shaping | |
| CN110674845A (en) | Dish identification method combining multi-receptive-field attention and characteristic recalibration |
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 | ||
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20190821 Address after: 350200 Changle City, Fujian Province, Wu Hang Zheng Helu, No., room 2, room 123, No. 601 Applicant after: Chen Qing Address before: 215300 No. 89 Zhangji Road, Small and Medium Enterprise Park, Kunshan Development Zone, Suzhou City, Jiangsu Province Applicant before: Sun Yan |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20221125 Address after: 276000 science and technology entrepreneurship Park b503-5, high tech Industrial Development Zone, Linyi City, Shandong Province Patentee after: Linyi zhidayun Intelligent Technology Co.,Ltd. Address before: 350200 Zheng Helu, 601, room 2, No. 123, Wu Hang, Changle, Fuzhou, Fujian Patentee before: Chen Qing |