CN105300895B - A kind of method using characteristic point tangent line angle early warning potato sprouting defect - Google Patents
A kind of method using characteristic point tangent line angle early warning potato sprouting defect Download PDFInfo
- Publication number
- CN105300895B CN105300895B CN201510747071.2A CN201510747071A CN105300895B CN 105300895 B CN105300895 B CN 105300895B CN 201510747071 A CN201510747071 A CN 201510747071A CN 105300895 B CN105300895 B CN 105300895B
- Authority
- CN
- China
- Prior art keywords
- early warning
- potato
- value
- point
- warning time
- 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
Landscapes
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The invention discloses a kind of method using characteristic point tangent line angle early warning potato sprouting defect.The high spectrum image of multiple sample potatos is gathered under the same conditions, record pre-warning time, choose the area-of-interest of potato, spectroscopic data is classified according to the different number of days of pre-warning time, spectroscopic data is intercepted after mean filter, build spectrum simulation function, derivation obtains first derivative figure, by minimum point, the included angle cosine value that maximum point and its point of intersection of tangents are formed is as characteristic value, discriminant analysis is carried out to obtain discriminant coefficient and differentiate constant, tested potato is repeated the above steps to obtain corresponding characteristic value, and discriminant analysis is carried out to it and obtains early warning result, realize the early warning to potato sprouting.The present invention realizes potato sprouting early warning using two wave bands, improves detection efficiency, reduces loss caused by process of circulation potato sprouting.
Description
Technical Field
The invention relates to a fruit and vegetable defect detection method, in particular to a method for early warning a potato germination defect by using a characteristic point tangent included angle.
Background
Potatoes are widely grown worldwide as one of four major food crops in the world. The potatoes have extremely high nutritional value, are reputed to be second bread and underground apples abroad, contain most of nutrients in grains, vegetables and fruits, and are basically possessed by the potatoes. Meanwhile, the potatoes have wide suitable planting regions, low requirements on soil moisture and fertility and huge yield increasing potential, and are considered by grain agriculture organization experts of the United nations to save human grain crops when grain crises occur in the future world.
During the processes of harvesting, storing, transporting and the like, the potatoes are easy to have various defects of mechanical damage, germ infection, green germination and the like, the quality of the potatoes is seriously influenced, and economic losses are brought to potato farmers and consumers. One of the national standard detection indexes of the potatoes is that the potato has no defects of frostbite, black heart, sprouting, green potato and the like. Meanwhile, germinated potatoes are toxic (Solanine), and the safety of people is threatened by accidental eating of the potatoes.
The research on the germination detection of potatoes is carried out in order to prevent the germination of the potatoes from entering the market due to improper storage or transportation, and also to lay a cushion for the automation of quality inspection and grading of the potatoes.
At present, a lot of achievements are made for detecting the external quality of potatoes at home and abroad. At present, detection aiming at the surface defects of potatoes is mainly concentrated on mechanical damage, holes, scabs, surface bruising, sprouting greening and the like, wherein in the aspect of potato sprouting detection, zheng Guannan and the like adopt a G-channel gray value difference method to detect the sprouting potatoes, and according to an experimental result, when the total number of bud points is more than 10, the existence of a sprouting body in an image can be basically judged.
The detection of the sprouted potatoes can be completed by utilizing an RGB color machine vision system and matching with a corresponding algorithm, and the domestic detection of the sprouted potatoes in the direction can reach higher accuracy at present. However, the method of sprouting detection based on the color characteristics of the potato surface cannot be used for predicting the sprouting time of the potatoes.
Disclosure of Invention
In order to solve the problems in the background art, the invention aims to provide a method for early warning the sprouting defect of potatoes by utilizing a characteristic point tangent included angle, wherein the region is classified according to the cosine value of a characteristic angle formed by spectral values of 3 wave bands of the region of interest, so that the sprouting early warning of potatoes is realized.
The technical scheme adopted by the invention for solving the technical problem is as follows:
1) Hyperspectral images of multiple sample potatoes were collected under the same conditions:
the step 1) is to collect the hyperspectral image by adopting the following method: taking black cardboard as background, fixing at least 150 potatoes on the cardboard by foam adhesive, and collecting hyperspectral images in a dark box every day for 5 days continuously.
2) Recording the days from the start of collecting the hyperspectral image to the sprouting of the potato as early warning time, selecting the sprouting part of the potato as an interested area, extracting the spectral data of the sprouting part, and classifying the spectral data according to different days of the early warning time; the specific implementation can be that the germination date of the potatoes is 0 day, and k days before germination are reversely deduced to k days before germination, so that k days are the early warning time;
in the step 2), the interesting area of the potato is selected in the following way: finding and recording the center (S, T) of the eye position of the germinated potato, and establishing a sprouting part area with the eye position (S, T) as the center and nine pixel points as the side length as an interested area for data processing.
3) And 3X 3 mean value filtering is carried out on all the wave band data of the region of interest, and the discrete spectrum data of the 600-750nm wave band where the region of interest is located are intercepted.
In the step 3), fitting the spectral data by using a spectral fitting function of the following formula, and solving by using an nlifit function in Matlab:
wherein, the abscissa x is the wavelength value, the ordinate f (x) is the spectrum uniformization value, n represents the accumulation parameter, and j represents the calculation ordinal number of the accumulation parameter.
In a specific implementation, the value of n is determined to be 5 using the root mean square error RMSE for evaluating the degree of fit.
4) Constructing a spectrum fitting function, deriving the spectrum fitting function f (χ) at different early warning times to obtain a first derivative graph of the region of interest, selecting a minimum value point A and a maximum value point C which are closest to the wavelength of 680nm from the first derivative graph by taking a wavelength value as a horizontal coordinate and a spectrum normalized value as a vertical coordinate, respectively making tangent lines at the minimum value point A and the maximum value point C, taking the intersection point of the two tangent lines as an intersection point B, and calculating a cosine value of an included angle ABC formed by the minimum value point A, the maximum value point C and the intersection point B as a characteristic value cosB;
the characteristic value cosB is specifically calculated in the following way: the coordinates of the minimum point A, the maximum point C and the intersection point B are respectively marked as (lambda) A ,R A )、(λ B ,R B ) And (lambda) C ,R C ) Wherein λ is A 、λ B And λ C Respectively corresponding to the minimum value point A, the maximum value point C and the intersection point B A 、R B And R C And calculating the included angle ABC by adopting the cosine law according to the coordinates of the three points.
5) Respectively carrying out discriminant analysis on the characteristic values cosB of the regions of interest at different early warning times to respectively obtain respective discriminant coefficients p k And a discrimination constant q k ；
In the step 5), the feature value cosB is subjected to discriminant analysis by using the following formula expressed by a Fisher discriminant coefficient discriminant method:
F 0 ＝p 0 ×X+q 0
F 1 ＝p 1 ×X+q 1
F 2 ＝p 2 ×X+q 2
F 3 ＝p 3 ×X+q 3
F 4 ＝p 4 ×X+q 4
wherein, F 0 Score value for potato sample at day0 early warning time, F 1 Score value for potato sample at day1 early warning time, F 2 Score value for potato sample at day2 early warning time, F 3 Score value for day 3 early warning time sample potatoes, F 4 The potato score value of the sample was the early warning time on day 4. X represents the set of characteristic values cos B at all warning times, p k A set of discrimination coefficients corresponding to the characteristic value cos B of the potato sample at the k-th early warning time, q k Is a discrimination constant;
6) And (5) repeating the steps 1) to 4) on the detected potatoes to obtain a corresponding characteristic value cosB, and carrying out discriminant analysis on the characteristic value cosB to obtain an early warning result, so that early warning of potato germination is realized.
The discrimination analysis result of the characteristic value cos B in the step 6) specifically adopts the following mode:
substituting all characteristic values cos B extracted from the detected potatoes into the following formula to obtain a score value F at each early warning time ks ，F ks K corresponding to the maximum value in the values is the early warning time:
F 0s ＝p 0 ×X s +q 0
F 1s ＝p 1 ×X s +q 1
F 2s ＝p 2 ×X s +q 2
F 3s ＝p 3 ×X s +q 3
F 4s ＝p 4 ×X s +q 4
wherein, F 0s Score of potato measured for day0 early warning time, F 1s Score for potatoes tested at day1 early warning time, F 2s Score of potato measured for day2 early warning time, F 3s Score for potatoes tested at day 3 early warning time, F 4s Score of potato measured for day4 early warning time, X s Representing a fitting parameter set of the detected potatoes;
the method continuously observes the surface bud eyes of the potatoes by utilizing a hyperspectral imaging technology, and records the bud eye condition until the potatoes germinate. And recording the germination date of the potatoes as 0 th day, reversely pushing the potatoes to k days before germination, wherein k is the early warning time, and counting and classifying the early warning time of each bud eye. And extracting the spectral data of a square area with 9 pixel points on the side length of the eye position of each potato bud, carrying out mean value normalization and function fitting on the spectrum of the 600-750nm wave band in the area after mean value filtering, and obtaining the spectral curve graphs of the eye positions of the potato buds under different early warning time. Using the first derivative of the curve as the wave band value lambda under the extreme value A 、λ C The tangent intersection B (lambda) is determined from the derivative value B ,R B ) And calculating cosine values of tangent included angles of all potato eyes under corresponding wave bands, and constructing a discrimination function by taking the cosine values as variables to realize early warning of potato germination.
The beneficial effects of the invention are:
the invention can obtain the early warning information of the germination of the potatoes, and can early warn the germination condition of the potatoes, thereby realizing the early warning of the germination of the potatoes, improving the detection efficiency and reducing the loss caused by the germination of the potatoes in the circulation process.
Drawings
FIG. 1 is a flow chart of an implementation of the method of the present invention.
FIG. 2 is a schematic view of a potato fixing condition.
Fig. 3 is an example of the extraction of the region of interest at the eye position of potato buds.
FIG. 4 is a graph of a spectral curve fit of the eye position warning time of potato buds with a value of 1 (day 1) in the wavelength range of 600-750nm in the examples.
FIG. 5 is a graph of the first derivative of the eye position warning time of potato buds with day1 in the 600-750nm band of the example.
Fig. 6 is a characteristic angle diagram of the early warning time of the potato bud eye position of 1 (day 1) in the embodiment.
FIG. 7 is a graph of the spectrum of potato buds at the eye site of germination day (day 0) in the wavelength range of 600-750nm in example.
FIG. 8 is a graph of first derivative of spectral curve of the potato bud at the day of eye germination (day 0) in the 600-750nm band of examples.
FIG. 9 is a schematic diagram of the characteristic angle B of the day of germination (day 0) at the eye position of potato sprout in example.
Fig. 10 is a schematic diagram of characteristic angle B with the early warning time of potato bud eye position of 2 (day 2) in the example.
Detailed Description
The invention is further illustrated by the following examples in conjunction with the drawings.
The embodiments of the invention are as follows:
as shown in FIG. 1, first, potatoes were fixed on a black cardboard (shown in FIG. 2), stored at room temperature in a dark place, and hyperspectral images thereof were collected every day. And debugging the hyperspectral imaging system, and matching the parameters of the hyperspectral imaging system, such as object distance, illumination intensity, camera exposure time, scanning area, scanning speed and the like so as to scan the hyperspectral image of the potato based on the clear and non-deformable image which can be collected. Eyes were recorded until sprouting. Recording the days from the start of collecting the hyperspectral image to the sprouting of the potato as early warning time, and selecting the sprouting part of the potato as an area of interestAnd the domain extracts the spectral data of the region, and classifies the spectral data according to different days of the early warning time. Extracting the spectral data of a square area with 9 pixel points on the side length of the eye position of each potato bud, performing mean value filtering, performing mean value normalization and function fitting on the spectrum of the 600-750nm wave band in the area, and obtaining the spectral curve graph of the potato bud at different early warning time. Using the first derivative of the curve as the wave band value lambda under the extreme value A 、λ C Calculating the band value lambda of the tangent included angle B B And calculating cosine values of characteristic angles of all potato eyes under corresponding wave bands, and constructing discriminant analysis by taking the cosine values as variables to realize early warning of potato germination.
The region of interest of the potato sprout eye is shown in fig. 3, and the main operation flow is as follows: and recording the positions (S, T) of the eyes of the germinated potatoes, and taking the (S, T) as the center and the area with 9 pixel points as the side length as the region of interest of data processing, wherein the area is defined as a germination part. And extracting the spectral data of the germination part, and classifying the spectral data according to the germination condition. The counted number of eye-bud parts under different early warning time is shown in table 1. day0 represents the day of germination, and day1-day4 represent the early warning time of 1-4 days.
TABLE 1 number of eye positions counted in different germination situations
FIG. 4 is a graph fitted with spectral data of early warning time of potato in 600-750nm wave band of 1. In the present invention, the fitting function format used is as follows:
the number of samples is 117 bud eye parts, and the spectrum discrete value is obtained by normalizing the mean value of 117 sample spectrum data.
FIG. 5 is a graph of the first derivative of the spectral curve of potato bud eye position early warning time 1 (day 1) in the 600-750nm band. Minimum point A (667.92, 0.038489) and maximum point C (688.16, 0.380951) nearest to the wavelength of 680nm are selected as feature points in the invention.
Fig. 6 is a schematic diagram of the characteristic angle B with the early warning time of the potato bud eye position of 1 (day 1). In the present invention, the included angle position B (678.03, 0.048108) is obtained using the first derivative value at the feature point a (667.92, 0.038489) and the feature point C (688.16, 0.380951). Determine lambda A 、λ B 、λ C After equal wavelength values, the ith potato sample (667.92 Ai )，B(678.03,R Bi )，C(688.16,R Ci ) The characteristic value cos B is calculated. (i =1,2,3 \8230117)
FIG. 7 is a graph showing the spectral profile of potato buds in the 600-750nm band at the day of eye germination (day 0). The number of samples is 118 bud eye parts, and the curve is obtained by normalizing the mean value of the spectrum data of 118 samples.
FIG. 8 is a graph showing the first derivative of the spectral curve of potato bud at the day of germination (day 0) in the eye position of potato buds in the 600-750nm band. Two extreme points A (667.92, 0.038489) and C (688.16, 0.380951) nearest to the wavelength of 680nm are selected as characteristic points in the invention.
Fig. 9 is a schematic diagram of characteristic angles B of the potato sprout eye position on the day of germination (day 0). In the present invention, A (667.92, 0.099426), B (678.03, 0.107984), C (688.16, 0.419124). Determine λ A 、λ B 、λ C After equal wavelength values, the ith potato sample (667.92 Ai )，B(678.03,R Bi )，C(688.16,R Ci ) The characteristic value cos B is calculated. (i =1,2,3 \ 8230118)
Fig. 10 is a schematic diagram of characteristic angle B of potato bud eye position early warning time of 2 (day 2). In the present invention, A (667.92, 0.103862), B (678.03, 0.111303), C (688.16, 0.420053). Determine λ A 、λ B 、λ C After an equal wavelength value, the ith potato sample (667.92 Ai )，B(678.03,R Bi )，C(688.16,R Ci ) The characteristic value cos B is calculated. (i =1,2,3 \8230110; 110)
And (3) discrimination analysis: and classifying the characteristic values cos B of the regions of interest at different early warning times by using a Fisher discrimination coefficient discrimination method provided by the SPSS. The discriminant function was established from a Karl Fischer discriminant function coefficient table in the processing results, as shown in Table 2.
TABLE 2 Fischer discriminant function coefficient table
The formula is thus calculated as follows:
F 0 ＝-66.898X-23.540
F 1 ＝-94.664X-45.523
F 2 ＝-92.748X-43.764
F 3 ＝-95.004X-45.839
F 4 ＝-90.524X-41.766
early warning of potato germination conditions: and early warning is carried out by using the obtained discrimination function. Substituting the cosine value of the tangent included angle of the feature point of the new sample as a variable into a discrimination function, and calculating to obtain various scores F ks And obtaining the maximum classification as an early warning classification. After the test, 118 samples on the germination day and 117 samples with the early warning time of 1 are taken, and the early warning results are shown in table 3, wherein 86.0% of the samples are correctly classified.
TABLE 3 early Potato Warning results
Therefore, the classification effect of the single characteristic value is improved, and a better result can be obtained for early warning of potato germination by combining other characteristic values.
Claims (6)
1. A method for early warning the germination defect of potatoes by utilizing a characteristic point tangent included angle is characterized by comprising the following steps:
1) Hyperspectral images of multiple sample potatoes were collected under the same conditions:
2) Recording the days from the start of collecting the hyperspectral image to the germination of the potato as early warning time, selecting the germination part of the potato as an interested area, extracting spectral data of the germination part, and classifying the spectral data according to different days of the early warning time;
3) Carrying out 3X 3 mean value filtering on all wave band data of the region of interest, and intercepting discrete spectrum data of a wave band of 600-750nm where the region of interest is located;
4) Constructing a spectrum fitting function, carrying out derivation on the spectrum fitting function f (x) with different early warning time to obtain a first derivative graph of the region of interest, selecting a minimum value point A and a maximum value point C which are closest to the wavelength of 680nm from the first derivative graph by taking a wavelength value as a horizontal coordinate and a spectrum normalized value as a vertical coordinate, respectively making tangent lines at the minimum value point A and the maximum value point C, taking the intersection point of the two tangent lines as an intersection point B, and calculating a cosine value of an included angle ABC formed by the minimum value point A, the maximum value point C and the intersection point B as a characteristic value cosB;
in the step 4), fitting the spectrum data by specifically adopting a spectrum fitting function of the following formula:
wherein, x is the wavelength value, f (x) is the uniform value of the spectrum, n represents the accumulation parameter, j represents the calculation ordinal number of the accumulation parameter;
5) Respectively carrying out discriminant analysis on the characteristic values cosB of the regions of interest at different early warning times to respectively obtain respective discriminant coefficients p k And a discrimination constant q k ；
6) And (3) repeating the steps 1) to 4) on the detected potatoes to obtain all characteristic values cosB, and carrying out discriminant analysis on the characteristic values cosB to obtain an early warning result, so that early warning of potato germination is realized.
2. The method for early warning of potato germination defects by using the included angle between the tangent lines of the characteristic points as claimed in claim 1, wherein: the step 1) is to collect the hyperspectral image by adopting the following method: fixing at least 150 potatoes on a black paperboard as a background, placing the black paperboard in a dark box, collecting hyperspectral images every day, and continuously collecting the hyperspectral images for 5 days.
3. The method for early warning of potato germination defects by using the included angle between the tangent lines at the characteristic points as claimed in claim 1, wherein the method comprises the following steps: in the step 2), the interesting area of the potato is selected in the following way: finding and recording the positions (S, T) of the eyes of the sprouted potatoes, and establishing a sprouting part area which takes the positions (S, T) of the eyes as the center and nine pixel points as the side length as an interested area.
4. The method for early warning of potato germination defects by using the included angle between the tangent lines of the characteristic points as claimed in claim 1, wherein: in the step 5), the feature value cosB is subjected to discriminant analysis by using the following formula expressed by a Fisher discriminant coefficient discriminant method:
F 0 ＝p 0 ×X+q 0
F 1 ＝p 1 ×X+q 1
F 2 ＝p 2 ×X+q 2
F 3 ＝p 3 ×X+q 3
F 4 ＝p 4 ×X+q 4
wherein, F 0 Score value for potato sample at day0 early warning time, F 1 Score value for potato sample at day1 early warning time, F 2 Score value for potato sample at day2 early warning time, F 3 Score value for potato sample at day 3 early warning time, F 4 Score for potato samples at day4 early warning time, X represents the set of characteristic values cosB at all early warning times, p k Early warning time sample potatoes for day kSet of discrimination coefficients corresponding to the potato's characteristic value cosB, q k Are the discrimination constants.
5. The method for early warning of potato germination defects by using the included angle between the tangent lines of the characteristic points as claimed in claim 1, wherein: the discrimination analysis results of all the characteristic values cosB in the step 6) specifically adopt the following modes: substituting all characteristic values cosB extracted from the detected potatoes into the following formula to obtain a score value F at each early warning time ks ，F ks K corresponding to the maximum value in the values is the early warning time:
F 0s ＝p 0 ×X s +q 0
F 1s ＝p 1 ×X s +q 1
F 2s ＝p 2 ×X s +q 2
F 3s ＝p 3 ×X s +q 3
F 4s ＝p 4 ×X s +q 4
wherein, F 0s Score for potato tested at day0 early warning time, F 1s Score for potatoes tested at day1 early warning time, F 2s Score of potato measured for day2 early warning time, F 3s Score of potatoes measured for day 3 early warning time, F 4s Score of potato measured for day4 early warning time, X s The set of fitting parameters for the potato under test is indicated.
6. The method for early warning of potato germination defects by using the included angle between the tangent lines at the characteristic points as claimed in claim 1, wherein the method comprises the following steps: the characteristic value cosB in the step 4) is calculated in the following way: the coordinates of the minimum point A, the maximum point C and the intersection point B are respectively marked as (lambda) A ,R A )、(λ B ,R B ) And (lambda) C ,R C ) Wherein λ is A 、λ B And λ C Respectively corresponding to the minimum value point A, the maximum value point C and the intersection point B, R A 、R B And R C Respectively calculating the included angle ABC for the spectrum uniformization values corresponding to the minimum value point A, the maximum value point C and the intersection point B by adopting the cosine theorem expressed by the following formula according to the coordinates of the three points:
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510747071.2A CN105300895B (en) | 2015-11-05 | 2015-11-05 | A kind of method using characteristic point tangent line angle early warning potato sprouting defect |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510747071.2A CN105300895B (en) | 2015-11-05 | 2015-11-05 | A kind of method using characteristic point tangent line angle early warning potato sprouting defect |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105300895A CN105300895A (en) | 2016-02-03 |
| CN105300895B true CN105300895B (en) | 2017-12-26 |
Family
ID=55198407
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510747071.2A Active CN105300895B (en) | 2015-11-05 | 2015-11-05 | A kind of method using characteristic point tangent line angle early warning potato sprouting defect |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105300895B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107451585B (en) * | 2017-06-21 | 2023-04-18 | 浙江大学 | Potato image recognition device and method based on laser imaging |
| CN109493287B (en) * | 2018-10-10 | 2022-03-15 | 浙江大学 | Deep learning-based quantitative spectral data analysis processing method |
| CN110047064B (en) * | 2019-03-27 | 2021-03-19 | 中国农业机械化科学研究院 | Potato scab detection method |
| CN112710661B (en) * | 2020-12-25 | 2022-07-22 | 河北北方学院 | Potato breeding monitoring and analyzing method and system |
| CN114820619B (en) * | 2022-06-29 | 2022-11-15 | 深圳市信润富联数字科技有限公司 | Tuber plant sorting method, system, computer device and storage medium |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6091843A (en) * | 1998-09-03 | 2000-07-18 | Greenvision Systems Ltd. | Method of calibration and real-time analysis of particulates |
| CN102313699A (en) * | 2011-05-26 | 2012-01-11 | 北京农业信息技术研究中心 | Estimation method of total nitrogen content in crop canopy leaf |
| CN102612892A (en) * | 2012-03-02 | 2012-08-01 | 北京农业信息技术研究中心 | Identification method for sprouting conditions of wheat ears |
| CN103336946A (en) * | 2013-06-17 | 2013-10-02 | 浙江大学 | Binocular stereoscopic vision based clustered tomato identification method |
| CN103630498A (en) * | 2013-11-12 | 2014-03-12 | 浙江大学 | Method for detecting pesticide residue on surface of navel orange based on hyperspectral imaging technology |
| CN204269559U (en) * | 2014-11-17 | 2015-04-15 | 浙江大学 | The complete surperficial high spectrum image online acquisition device of a kind of elliposoidal fruit |
| CN104597052A (en) * | 2015-02-09 | 2015-05-06 | 淮阴工学院 | High-speed lossless potato grading detection method and system based on multi-characteristic fusion |
| CN105004737A (en) * | 2015-07-14 | 2015-10-28 | 浙江大学 | Self-adaption improved gradient information-based fruit surface defect detection method |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB2397423B (en) * | 2001-09-17 | 2005-06-01 | Ca Minister Agriculture & Food | A method and apparatus for identifying and quantifying characteristics of seeds and other small objects |
| JP4239955B2 (en) * | 2004-11-19 | 2009-03-18 | コニカミノルタセンシング株式会社 | Multi-angle colorimeter, lighting device and light receiving device |
| JP5021293B2 (en) * | 2006-12-29 | 2012-09-05 | 株式会社パスコ | Crop growth status analysis method, crop growth status analysis device, and crop growth status analysis program |
| JP2013169156A (en) * | 2012-02-18 | 2013-09-02 | Mie Univ | Color image pickup and appearance feature analysis system |
-
2015
- 2015-11-05 CN CN201510747071.2A patent/CN105300895B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6091843A (en) * | 1998-09-03 | 2000-07-18 | Greenvision Systems Ltd. | Method of calibration and real-time analysis of particulates |
| CN102313699A (en) * | 2011-05-26 | 2012-01-11 | 北京农业信息技术研究中心 | Estimation method of total nitrogen content in crop canopy leaf |
| CN102612892A (en) * | 2012-03-02 | 2012-08-01 | 北京农业信息技术研究中心 | Identification method for sprouting conditions of wheat ears |
| CN103336946A (en) * | 2013-06-17 | 2013-10-02 | 浙江大学 | Binocular stereoscopic vision based clustered tomato identification method |
| CN103630498A (en) * | 2013-11-12 | 2014-03-12 | 浙江大学 | Method for detecting pesticide residue on surface of navel orange based on hyperspectral imaging technology |
| CN204269559U (en) * | 2014-11-17 | 2015-04-15 | 浙江大学 | The complete surperficial high spectrum image online acquisition device of a kind of elliposoidal fruit |
| CN104597052A (en) * | 2015-02-09 | 2015-05-06 | 淮阴工学院 | High-speed lossless potato grading detection method and system based on multi-characteristic fusion |
| CN105004737A (en) * | 2015-07-14 | 2015-10-28 | 浙江大学 | Self-adaption improved gradient information-based fruit surface defect detection method |
Non-Patent Citations (9)
| Title |
|---|
| Automatic detection of common surface defects on oranges using combined lighting transform and image ratio methods;Jiangbo Li.et al;《Postharvest Biology and Technology》;20131231;第82卷;第59-69页 * |
| Detection of common defects on oranges using hyperspectral reflectance imaging;Jiangbo Li.et al;《Computers and Electronics in Agriculture》;20110529;第78卷;第38-48页 * |
| GERMINATION OF SEEDS AS INFLUENCED BY MATRIC POTENTIAL AND BY AREA OF CONTACT BETWEEN SEED AND SOIL WATER;N. COLLIS-GEORGE.et al;《Soil Research》;19661231;第4卷(第2期);第145-164页 * |
| The application of machine vision to food and agriculture: a review;E R Davies.et al;《The Imaging Science Journal》;20130718;第197-217页 * |
| 基于高光谱图像和光谱信息融合的马铃薯多指标检测方法;金瑞等;《农业工程学报》;20150831;第31卷(第16期);第258-263页 * |
| 基于高光谱成像技术检测脐橙溃疡;李江波等;《农业工程学报》;20100831;第26卷(第8期);第222-228页 * |
| 基于高光谱成像技术的马铃薯外部缺陷检测;周竹等;《农业工程学报》;20121130;第28卷(第21期);第221-228页 * |
| 基于高光谱漫透射成像整体检测番茄可溶性固形物含量;张若宇等;《农业工程学报》;20131231;第29卷(第23期);第247-252页 * |
| 高光谱图像技术结合图像处理方法检测马铃薯外部缺陷;苏文浩等;《浙江大学学报（农业与生命科学版）》;20131205;第40卷(第2期);第188-196页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105300895A (en) | 2016-02-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105300895B (en) | A kind of method using characteristic point tangent line angle early warning potato sprouting defect | |
| Bhargava et al. | Fruits and vegetables quality evaluation using computer vision: A review | |
| Mahendran et al. | Application of computer vision technique on sorting and grading of fruits and vegetables | |
| Dey et al. | Image processing based leaf rot disease, detection of betel vine (Piper BetleL.) | |
| Zhou et al. | Diagnosis of winter-wheat water stress based on UAV-borne multispectral image texture and vegetation indices | |
| Cubero et al. | Advances in machine vision applications for automatic inspection and quality evaluation of fruits and vegetables | |
| Duan et al. | Remote estimation of grain yield based on UAV data in different rice cultivars under contrasting climatic zone | |
| Sethy et al. | Hyperspectral imagery applications for precision agriculture-a systemic survey | |
| CN102612892B (en) | Identification method for sprouting conditions of wheat ears | |
| CN105426585B (en) | A kind of potato sprouting method for early warning based on SIN function fitting process | |
| Zhang et al. | Detection and classification of potato defects using multispectral imaging system based on single shot method | |
| CN104297165A (en) | Hyperspectral image-based method for predicting growth of rot funguses | |
| CN105067532A (en) | Method for identifying early-stage disease spots of sclerotinia sclerotiorum and botrytis of rape | |
| CN115170958A (en) | Method for calculating wilting degree of soybean canopy by adopting multi-spectral image Fourier transform | |
| Torres et al. | Setting up a methodology to distinguish between green oranges and leaves using hyperspectral imaging | |
| CN105424622B (en) | A kind of method using feature triangle area early warning potato sprouting | |
| Ghorai et al. | Image processing based detection of diseases and nutrient deficiencies in plants | |
| Zhao et al. | Early detection of aphid (Myzus persicae) infestation on Chinese cabbage by hyperspectral imaging and feature extraction | |
| CN114199880A (en) | Citrus disease and insect pest real-time detection method based on edge calculation | |
| Khadabadi et al. | Disease detection in vegetables using image processing techniques: A review | |
| Xu et al. | A deep learning model for rapid classification of tea coal disease | |
| Mahmud et al. | Detection of strawberry powdery mildew disease in leaf using image texture and supervised classifiers | |
| Singh et al. | A novel algorithm for segmentation of diseased apple leaf images | |
| Mohamed et al. | Strawberry ripeness calibrated 2D colour lookup table for field-deployable computer vision | |
| Ali et al. | Potential use of digital photographic technique to examine wild blueberry ripening in relation to time of harvest |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant |