CN104237235A - Rapid detection method based on near-infrared imaging technology for food-borne pathogens - Google Patents
Rapid detection method based on near-infrared imaging technology for food-borne pathogens Download PDFInfo
- Publication number
- CN104237235A CN104237235A CN201410464917.7A CN201410464917A CN104237235A CN 104237235 A CN104237235 A CN 104237235A CN 201410464917 A CN201410464917 A CN 201410464917A CN 104237235 A CN104237235 A CN 104237235A
- Authority
- CN
- China
- Prior art keywords
- food
- borne pathogens
- infrared imaging
- detection method
- imaging technology
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Landscapes
- Measuring Or Testing Involving Enzymes Or Micro-Organisms (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The invention relates to a rapid detection method for food-borne pathogens and particularly relates to a rapid detection method based on a near-infrared imaging technology for food-borne pathogens, belonging to the field of microbiological detection. The rapid detection method comprises the following steps: acquiring texture images formed by optical transmission and scattering phenomena generated in different bacterial colonies by virtue of a near-infrared imaging system, wherein the images can reflect internal structure feature information of the bacterial colonies; then performing normalization preprocessing on the texture images by virtue of standard torque to endow the preprocessed images with translation and scale invariance, and extracting a Z-torque feature with rotation invariance from the processed images; and finally, establishing a classification model of the different bacterial colonies, and identifying bacterial types of similar unknown bacterial colony samples. The images acquired by the detection method can reflect the internal structure features of the bacterial colonies, and compared with the prior art, the detection method has the advantages of no complicated sample preprocessing, high detection speed, convenience and simplicity in operation and high identification accuracy.
Description
Technical field
The present invention relates to a kind of method for quick for food-borne pathogens, refer in particular to the method for quick of the food-borne pathogens based on near-infrared imaging technology, belong to microorganism detection field.
Background technology
Food-borne pathogens is the first cause causing food origin disease, and the whole world occurs every year up in the diarrhoea case of 1,500,000,000, causes about less than 3,000,000 5 years old death of child, and wherein the case of 70% is caused by the food owing to being subject to biogenic pollution.In developing country, the case nearly 2.7 hundred million of annual diarrhoea and other relevant food origin disease, what wherein cause less than 5 years old death of child has 2,400,000 examples.At present, domestic common food-borne pathogens mainly contains Escherichia coli, staphylococcus aureus and salmonella etc.Therefore, strictly carrying out Food Safety Analysis is reduce the important channel of food origin disease, and fast, easy, special detection method becomes study hotspot.
In Food microbe testing, there are traditional immunological method, the immunological method of improvement, and the biochip technology and PCR technology etc. of to get up along with Development of Molecular Biology, high and the detection speed of the sensitivity of these methods also improves a lot, but complicated operation, expensive reagents, needs the operating personnel of specialty.In recent years due to the fast development of computer technology, microorganism detection method starts the future development to seriation, milligram ammonia and robotization, and a series of detecting instrument starts to be widely used in Microbiological detection of foods field.At present, the main method detecting microorganism fast has chromatographic technique, luciferase mark immunoassay, fourier transform infrared spectroscopy and Polymerize chain reaction technology etc.But the false positive reaction still existed in these new methods in various degree, still need to combine the conventional method of inspection.
Therefore, study fast and accurately that food-borne pathogens detection method is extremely important.
Summary of the invention
The object of the invention is for the deficiencies in the prior art, a kind of method can carrying out detection fast to food-borne pathogens is provided.
The present invention is by the texture image of near infrared imaging system acquisition bacterial clump sample, and this image can reflect the inner structural features of bacterium colony.First standard square is used to be normalized pre-service to original image, pretreated image is made to have translation and scale invariance, then therefrom extract the Z-moment characteristics with rotational invariance, finally adopt chemometrics method to set up the taxonomic history model of bacterial clump.To similar sample to be tested by corresponding image acquisition and feature extraction, then differentiate this specimen types by the corresponding model set up.
The present invention realizes by the following method:
First, choose several frequently seen food-borne pathogens (Escherichia coli, staphylococcus aureus and salmonella etc.) spread plate and cultivate a period of time, to often kind of bacterium, choose bacterium colony that on multiple flat board, size is basically identical as sample.Then, by the texture image of near infrared imaging system acquisition sample, standard square is utilized to be normalized pre-service to original image, pretreated image is made to have translation and scale invariance, then the Z-moment characteristics with rotational invariance is therefrom extracted, using the mould of Z-square as the characteristic variable of sample.Finally utilize the taxonomic history model of the method establishment bacterial clump of back-propagation artificial neural network, and use institute's established model to realize the discriminating of similar unknown sample.
When set up model observation effect is assessed, random selecting tested bacteria bacterium colony sample 2/3 as training group, for the foundation of model, remaining 1/3 as checking group, for the inspection of model prediction result, relative discern rate, determines best forecast model.
The method for quick of the food-borne pathogens based on near-infrared imaging technology of the present invention, concrete operations are as following steps:
(1) preparation of bacterial clump sample: choose several frequently seen food-borne pathogens (Escherichia coli, staphylococcus aureus and salmonella etc.) spread plate and cultivate;
(2) collection of near infrared texture image: the generating laser selecting near-infrared band, vertical irradiation bacterial clump, utilizes near infrared camera to gather the texture image of bacterium colony sample, as the original image of this bacterium colony sample;
(3) pre-service of original image and feature extraction: first use standard square to be normalized pre-service to original image, then extract Z-moment characteristics, using the mould of Z-square as characteristic variable;
(4) to classify the foundation of forecast model and evaluation: adopt chemometrics method to set up the disaggregated model of bacterial clump, and use the classification of the similar unknown bacterium colony sample of this model realization.
In step (2), described near infrared imaging system comprises light box, near infrared camera, generating laser, diffuse reflection screen and computing machine, wherein near infrared camera, generating laser and diffuse reflection screen are positioned at light box, and near infrared camera is connected with computing machine by USB data transmission line.The center emission wavelength of generating laser is 980 nm, and selects the industrial CCD camera within the scope of near-infrared band to carry out the collection of bacterium texture image.This method is characterised in that the texture image of gathered near-infrared region, the texture image that the generating laser cannot observed for human eye is formed through bacterial clump generation optical transmission and scattering phenomenon.
In step (2), the collection of described near infrared texture image is as follows: be irradiated to the bacterium colony on flat board when laser vertical, part light directly penetrates bacterium colony and is transmitted to diffuse reflection screen, another part light is radiated at and optical scattering phenomenon occurs bacterium colony and projects diffuse reflection screen, the texture image that transmitted light and scattered light are formed on this screen, by near infrared camera collect and transmit to computing machine.
In step (3), use standard square to be normalized pre-service to original image, make pretreated image have translation and scale invariance, thus eliminate the bacterium colony impact caused not of uniform size.Described standard square is defined as:
Described standard square is defined as
Wherein, p, q are positive integer or 0;
for input picture.
In step (3), by pretreated for normalization image zooming-out Z-moment characteristics, using the mould of Z-square as characteristic variable, finally adopt the taxonomic history model of the method establishment bacterial clump of back-propagation artificial neural network.Described Z-square formula is as follows:
Wherein, Znm is the rotation invariant moment of Z-square;
for input picture;
for polynomial expression
conjugation;
represent that initial point is to pixel
the distance of vector;
represent vector
follow
angle (counterclockwise) between axle; N, m are positive integer, and meet
for even number.
Z-square can any configuration High Order Moment, and High Order Moment comprises more picture information.What low-order moment proper vector described is the global shape of piece image target, and High Order Moment proper vector describes is the details of image object.
The wavelength response range of described near infrared camera is 900-1700nm, and described generating laser is the pointolite of about 980nm, and camera lens is to the distance of diffuse reflection screen and generating laser to dull and stereotyped distance experimentally condition and determining.
Beneficial effect of the present invention:
The method for quick of the food-borne pathogens based on near-infrared imaging technology provided by the invention, the image obtained can reflect the inner structural features of bacterial clump.Compared with prior art, sample is without the need to the pre-treatment of complexity, and detection speed is fast, simple to operation, differentiates that accuracy rate is high.
Accompanying drawing explanation
Fig. 1 is the near infrared imaging system schematic of embodiment;
Fig. 2 is the texture image of three kinds of bacterial clumps that embodiment gathers;
Fig. 3 is the taxonomic history result that embodiment adopts back-propagation artificial neural network method.
Embodiment
The quick detection of the present invention to bacterial clump has versatility, but because bacterial species is more, the present invention only have chosen three kinds of main food-borne pathogens (Escherichia coli, staphylococcus aureus and salmonella) and is detected as embodiment, and the detection of other bacteriums can with reference to the method for this embodiment.
Example implementation step is described in detail by reference to the accompanying drawings.
(1) preparation of test sample
Be coated with dull and stereotyped after the bacterium liquid of Escherichia coli, staphylococcus aureus and salmonella being carried out gradient dilution, in 37 DEG C of incubators, cultivate 18 ~ 24h, the bacterial clump on flat board is controlled between 50 ~ 80.Often kind of bacterium is coated with 5 flat boards and cultivates, and finally chooses the basically identical bacterium colony of 30 sizes as sample from flat board.
(2) collection of near infrared texture image
The structure of near-infrared laser scatter imaging system as shown in Figure 1, comprises light box, near infrared camera, generating laser, diffuse reflection screen and computing machine.During collection, for avoiding texture image to be out of shape, laser vertical is aimed at the center of dull and stereotyped upper bacterium colony, and generating laser keeps constant (30 mm) with dull and stereotyped distance, centre wavelength is about 980 nm.The wavelength response range of near infrared camera is 900-1700nm, collects the texture image of 320 × 256 pixel, as shown in Figure 2.
(3) pre-service of texture image and feature extraction
First utilize standard square to be normalized pre-service to original image, make pretreated image have translation and scale invariance, then therefrom extract the Z-moment characteristics with rotational invariance, using the mould of Z-square as the characteristic variable of sample.What low-order moment proper vector described is the global shape of piece image, and High Order Moment proper vector describes is the details of image.In the present embodiment, the exponent number n of Z-square is maximum is taken as 20, then altogether obtain 90 characteristic variables.
(4) foundation of forecast model and evaluation
All samples are divided into training set and forecast set according to 2:1, then training set has 60 bacterium colony samples for the structure of model, and forecast set has 30 samples for the checking of model.Back-propagation artificial neural network (BPANN) is utilized to build the disaggregated model of bacterium colony sample after feature extraction.As shown in Figure 3, when number of principal components is 6, the discrimination of BPANN model is the highest for result, and training set is 100%, and forecast set is 90%.Result shows, utilizes method of the present invention to may be used for the quick detection of food-borne pathogens.
Claims (6)
1., based on a method for quick for the food-borne pathogens of near-infrared imaging technology, comprise the following steps:
The preparation of bacterial clump sample: choose several frequently seen food-borne pathogens, as the spread plates such as Escherichia coli, staphylococcus aureus and salmonella are cultivated;
The collection of near infrared texture image: the generating laser selecting near-infrared band, vertical irradiation bacterial clump, utilizes near infrared camera to gather the texture image of bacterium colony sample, as the original image of this bacterium colony sample;
The pre-service of original image and feature extraction: first use standard square to be normalized pre-service to original image, then extract Z-moment characteristics, using the mould of Z-square as characteristic variable;
The foundation of classification forecast model and evaluation: adopt chemometrics method to set up the disaggregated model of bacterial clump, and use the classification of the similar unknown bacterium colony sample of this model realization.
2. the method for quick of a kind of food-borne pathogens based on near-infrared imaging technology according to claim 1, it is characterized in that in step (2), described near infrared imaging system comprises light box, near infrared camera, generating laser, diffuse reflection screen and computing machine, wherein near infrared camera, generating laser and diffuse reflection screen are positioned at light box, and near infrared camera is connected with computing machine by USB data transmission line; The center emission wavelength of generating laser is 980 nm.
3. the method for quick of a kind of food-borne pathogens based on near-infrared imaging technology according to claim 1, it is characterized in that in step (2), the collection of described near infrared texture image is as follows: be irradiated to the bacterium colony on flat board when laser vertical, part light directly penetrates bacterium colony and is transmitted to diffuse reflection screen, another part light is radiated at and optical scattering phenomenon occurs bacterium colony and projects diffuse reflection screen, the texture image that transmitted light and scattered light are formed on this screen, by near infrared camera collect and transmit to computing machine.
4. the method for quick of a kind of food-borne pathogens based on near-infrared imaging technology according to claim 1, is characterized in that, in step (3), described standard square is defined as
Wherein, p, q are positive integer or 0;
for input picture.
5. the method for quick of a kind of food-borne pathogens based on near-infrared imaging technology according to claim 1, is characterized in that in step (3), and described Z-square formula is as follows:
Wherein, Znm is the rotation invariant moment of Z-square;
for input picture;
for polynomial expression
conjugation;
represent that initial point is to pixel
the distance of vector;
represent vector
follow
angle (counterclockwise) between axle; N, m are positive integer, and meet
for even number.
6. the method for quick of a kind of food-borne pathogens based on near-infrared imaging technology according to claim 2, it is characterized in that, the wavelength response range of described near infrared camera is 900-1700nm, described generating laser is the pointolite of about 980nm, and camera lens is to the distance of diffuse reflection screen and generating laser to dull and stereotyped distance experimentally condition and determining.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410464917.7A CN104237235A (en) | 2014-09-15 | 2014-09-15 | Rapid detection method based on near-infrared imaging technology for food-borne pathogens |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410464917.7A CN104237235A (en) | 2014-09-15 | 2014-09-15 | Rapid detection method based on near-infrared imaging technology for food-borne pathogens |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN104237235A true CN104237235A (en) | 2014-12-24 |
Family
ID=52225740
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410464917.7A Pending CN104237235A (en) | 2014-09-15 | 2014-09-15 | Rapid detection method based on near-infrared imaging technology for food-borne pathogens |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104237235A (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105424645A (en) * | 2015-10-29 | 2016-03-23 | 福州大学 | Method for fast identifying clinical pathogens based on principal component analysis and Fisher discriminance |
| CN105651679A (en) * | 2016-02-04 | 2016-06-08 | 华中农业大学 | Method for quickly classifying bacterial colonies on culture medium on basis of hyperspectral imaging technology |
| CN106086155A (en) * | 2016-06-27 | 2016-11-09 | 浙江省农业科学院 | Strawberry surface Salmonella typhimurium forecast model and method for building up thereof under room temperature condition |
| CN107358193A (en) * | 2017-07-07 | 2017-11-17 | 南京天数信息科技有限公司 | Dermatophyte recognition detection method based on InceptionV3+ fully-connected networks |
| CN107580715A (en) * | 2015-04-23 | 2018-01-12 | Bd科斯特公司 | Method and system for automatic counting microbial colonies |
| CN110196251A (en) * | 2019-06-13 | 2019-09-03 | 六盘水市食品药品检验检测所 | A kind of sampling system and method for food inspection |
| CN112611830A (en) * | 2020-11-30 | 2021-04-06 | 湖北文理学院 | Method for distinguishing varieties of walnuts according to oxidation characteristics of walnuts |
| CN113447457A (en) * | 2021-01-22 | 2021-09-28 | 广东中烟工业有限责任公司 | Method for rapidly identifying optimal mould variety of mildewed tobacco |
-
2014
- 2014-09-15 CN CN201410464917.7A patent/CN104237235A/en active Pending
Non-Patent Citations (5)
| Title |
|---|
| ALIREZA KHOTANZAD等: "Invariant image recognition by Zernike moments", 《IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE》 * |
| BULENT BAYRAKTAR等: "feature extraction from light-scatter patterns of Listeria colonies for identification and classfication", 《JOURNAL OF BIOMEDICAL OPTICS》 * |
| EUIWON BAE等: ""Investigation of the presence of rod-shaped bacteria on food surface via elastic light scattering", 《ADVANCES IN BIOSCIENCE AND BIOTECHNOLOGY》 * |
| WAMIQ M.AHMED等: "classification of bacterial contamination using image processing and distributed computing", 《IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS》 * |
| 彭彦昆等: "光谱技术在生鲜肉品质安全快速检测的研究进展", 《食品安全质量检测学报》 * |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107580715A (en) * | 2015-04-23 | 2018-01-12 | Bd科斯特公司 | Method and system for automatic counting microbial colonies |
| CN107580715B (en) * | 2015-04-23 | 2022-05-31 | Bd科斯特公司 | Method and system for automatically counting microbial colonies |
| US11674116B2 (en) | 2015-04-23 | 2023-06-13 | Bd Kiestra B.V. | Method and system for automated microbial colony counting from streaked sample on plated media |
| CN105424645A (en) * | 2015-10-29 | 2016-03-23 | 福州大学 | Method for fast identifying clinical pathogens based on principal component analysis and Fisher discriminance |
| CN105651679A (en) * | 2016-02-04 | 2016-06-08 | 华中农业大学 | Method for quickly classifying bacterial colonies on culture medium on basis of hyperspectral imaging technology |
| CN106086155A (en) * | 2016-06-27 | 2016-11-09 | 浙江省农业科学院 | Strawberry surface Salmonella typhimurium forecast model and method for building up thereof under room temperature condition |
| CN106086155B (en) * | 2016-06-27 | 2020-01-17 | 浙江省农业科学院 | Strawberry surface salmonella typhimurium prediction model under room temperature condition and establishment method thereof |
| CN107358193A (en) * | 2017-07-07 | 2017-11-17 | 南京天数信息科技有限公司 | Dermatophyte recognition detection method based on InceptionV3+ fully-connected networks |
| CN110196251A (en) * | 2019-06-13 | 2019-09-03 | 六盘水市食品药品检验检测所 | A kind of sampling system and method for food inspection |
| CN112611830A (en) * | 2020-11-30 | 2021-04-06 | 湖北文理学院 | Method for distinguishing varieties of walnuts according to oxidation characteristics of walnuts |
| CN113447457A (en) * | 2021-01-22 | 2021-09-28 | 广东中烟工业有限责任公司 | Method for rapidly identifying optimal mould variety of mildewed tobacco |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104237235A (en) | Rapid detection method based on near-infrared imaging technology for food-borne pathogens | |
| He et al. | Application of deep learning in integrated pest management: a real‐time system for detection and diagnosis of oilseed rape pests | |
| US8787633B2 (en) | System and method of organism identification | |
| CN102564964B (en) | Spectral image-based meat quality visual non-contact detection method | |
| Lin et al. | The identification and filtering of fertilized eggs with a thermal imaging system | |
| CN104316473A (en) | Gender determination method for chicken hatching egg incubation early embryo based on hyperspectral image | |
| CN103900972B (en) | Multi-feature fusion-based meat freshness hyperspectral image visual detection | |
| CN103439342A (en) | Infrared nondestructive testing method based on thermal image time sequence characteristics | |
| CN106018332A (en) | Near-infrared-spectrum citrus yellow shoot disease field detection method | |
| Lee et al. | Identification of tea foliar diseases and pest damage under practical field conditions using a convolutional neural network | |
| JP2012135240A (en) | Apparatus and method for identifying bacterial colony | |
| JP6830593B2 (en) | How to identify microorganisms | |
| CN108985237A (en) | A kind of detection method and its system of the wheat scab based on depth mixing | |
| CN108663367A (en) | A kind of egg quality lossless detection method based on egg unit weight | |
| US20080102487A1 (en) | Method and apparatus for non-invasive rapid fungal specie (mold) identification having hyperspectral imagery | |
| CN104198324A (en) | Computer vision-based method for measuring proportion of cut leaves in cut tobacco | |
| Zhihui et al. | Nondestructive detection of infertile hatching eggs based on spectral and imaging information | |
| Liu et al. | Rapid identification of chrysanthemum teas by computer vision and deep learning | |
| Saifullah | K-means clustering for egg embryo’s detection based-on statistical feature extraction approach of candling eggs image | |
| Eady et al. | Methods for hyperspectral microscope calibration and spectra normalization from images of bacteria cells | |
| CN104297136A (en) | Hyperspectral image-based method for forecasting growth of pseudomonas aeruginosa | |
| CN109342378B (en) | Colony growth state detection device and method based on multi-modal imaging technology | |
| CN107735838A (en) | It is used for the abnormality detection of medical sample under a variety of settings | |
| Zhu et al. | Development of a novel noninvasive quantitative method to monitor Siraitia grosvenorii cell growth and browning degree using an integrated computer‐aided vision technology and machine learning | |
| KR101907563B1 (en) | Method for the viability of embryo in white fertilized egg using hyperspectral imaging |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20141224 |