CN113378882A - Method and system for rapidly detecting nutrient content in dish food by fusing maps - Google Patents
Method and system for rapidly detecting nutrient content in dish food by fusing maps Download PDFInfo
- Publication number
- CN113378882A CN113378882A CN202110512979.0A CN202110512979A CN113378882A CN 113378882 A CN113378882 A CN 113378882A CN 202110512979 A CN202110512979 A CN 202110512979A CN 113378882 A CN113378882 A CN 113378882A
- Authority
- CN
- China
- Prior art keywords
- sample
- target
- nutrient content
- detected
- spectral
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The invention provides a method and a system for rapidly detecting the nutrient content in dishes by fusing maps, wherein the method comprises the following steps: the method comprises the following steps of (1) stirring dish food and filtering liquid to form a target to be detected, and placing the target to be detected in a first container; acquiring image information of a target to be detected, and acquiring an estimated volume of the target to be detected according to the image information; acquiring spectral information of a target to be detected, and inputting the spectral information into a spectral density model to obtain the estimated density of the target to be detected; and calculating to obtain the estimated mass of the target to be detected according to the estimated volume and the estimated density, and inputting the estimated mass and the spectral information of the target to be detected into a spectral nutrient content model of the target to be detected to obtain the nutrient content of the target to be detected. The method realizes the dish food type identification, the establishment of the image and dish food volume and density model and the establishment of the spectrum and dish food nutrient content model by utilizing the optical map technology and the machine learning and data modeling algorithm, and further estimates the quality and the nutrient content of the dish food according to the volume.
Description
Technical Field
The invention relates to the technical field of food detection, in particular to a method and a system for rapidly detecting the nutrient content in dishes and food by fusing a map.
Background
Diet is the material basis for human health, survival and longevity. With the improvement of living standard of people, the accurate nutritional food becomes a new demand growth point of people for food. The dish is an indispensable food in daily dish food of people, the insufficient or excessive nutrient intake of the daily dish food has important influence on human health, particularly, the nutrient content of the ingested dish food needs to be detected in real time for diabetes patients, hypertension patients, obesity, special people and the like, the dish form and the component composition are complex, the method for detecting the nutrient content in the conventional dish food mostly adopts physicochemical tests, the method is troublesome and labor-consuming, the cost is high, the method is not convenient to use and popularize, and the method also becomes a great obstacle for influencing people to realize accurate diet. The spectrum technology is widely applied to the field of food rapid detection. Scholars at home and abroad obtain information such as images, volumes and the like of dishes of a meal by combining an image technology with a machine learning algorithm, and then conjecture the nutritional content of the dishes by combining a dish food database so as to infer the calorie content value. The method does not consider factors such as different proportions and different cooking conditions among dishes, and the obtained dishes have inaccurate nutrient content. The spectrum technology can measure the content in a sample, and a student detects the minced dish food by utilizing the near infrared spectrum technology to obtain the calorie value of the unit weight of the dish food in the fixed container, but the method cannot directly obtain the total content of the nutrient content in the dish food, and before detection, the model matching can be completed by selecting the type of the dish food, and the method cannot be widely applied in practice.
Disclosure of Invention
The invention provides a method for rapidly detecting the nutrient content in dish food by fusing maps, which is used for solving the defect that the nutrient content of the dish food cannot be accurately detected in the prior art, and realizes the identification of the type of the dish food, the establishment of an image and dish food volume model, the establishment of a spectrum and dish food density model and the establishment of a spectrum and dish food nutrient content model by utilizing an optical map technology, machine learning and a data modeling algorithm, so that the quality and the nutrient content of the dish food are comprehensively estimated according to the volume of the dish.
The invention also provides a system for intelligently measuring the nutrient content of the dish food, which is used for solving the defect that the nutrient content of the dish food cannot be accurately detected in the prior art, and realizes the identification of the type of the dish food, the establishment of an image and dish food volume model, the establishment of a spectrum and dish food density model and the establishment of a spectrum and dish food nutrient content model by utilizing an optical spectrum technology, machine learning and a data modeling algorithm, so that the quality and the nutrient content of the dish food are comprehensively estimated according to the volume of the dish food.
According to the first aspect of the invention, the method for rapidly detecting the nutrient content in the dish food by using the fusion map comprises the following steps:
the method comprises the steps of stirring dish food and filtering liquid to form a target to be detected, and placing the target to be detected in a first container;
acquiring image information of the target to be detected, and acquiring the estimated volume of the target to be detected according to the image information;
acquiring spectral information of the target to be detected, and inputting the spectral information into a spectral density model to obtain the estimated density of the target to be detected, wherein the spectral density model is obtained by training based on sample spectral information of a sample and sample density corresponding to the sample spectral information;
calculating to obtain the estimated mass of the target to be detected according to the estimated volume and the estimated density, and inputting the estimated mass and the spectral information of the target to be detected into a spectral nutrient content model of the target to be detected to obtain the nutrient content of the target to be detected, wherein the spectral nutrient content model is obtained based on sample spectral information of a sample and sample nutrient content training corresponding to the sample spectral information.
According to an embodiment of the present invention, the step of obtaining the image information of the object to be measured and obtaining the estimated volume of the object to be measured according to the image information includes:
collecting image information of the target to be detected after the first container is stood;
acquiring a point at the edge of an image in image information as a first pixel point, and calculating to obtain a first estimated volume of the target to be measured according to the distance from the first pixel point to the edge of the first container, the distance from the first pixel point to the center of the image, the depth of the first container and the bottom radius of the first container, wherein the first estimated volume is the volume of the target to be measured based on the first pixel point;
repeating the steps until N estimated volumes of N pixel points are obtained, wherein N is more than or equal to two;
and calculating to obtain a first average value according to the N estimated volumes, and taking the first average value as the estimated volume.
Specifically, the embodiment provides an implementation method for estimating an estimated volume of a target to be measured according to image information, the method includes acquiring an image of the target to be measured by collecting the image of the target to be measured, extracting a pixel point on the target to be measured in the image, estimating the volume of the target to be measured according to the pixel point, and obtaining an estimated volume approximate to the target to be measured by averaging after a plurality of pixel points are selected for measurement to obtain a plurality of estimated volumes.
According to an embodiment of the present invention, the step of calculating a first average value according to the N estimated volumes and using the first average value as the estimated volume specifically includes:
acquiring a preset deviation and establishing N normal distribution functions of the pre-estimated volume by taking the preset deviation as a gradient interval;
and obtaining M estimated volumes within a normal distribution confidence interval, and calculating to obtain the first average value according to the M estimated volumes, wherein M is less than N.
Specifically, this embodiment provides another implementation manner for estimating the estimated volume of the target to be measured according to the image information, in order to obtain an estimated volume closer to the actual volume of the target to be measured, a functional relationship graph of each pixel point in the image and the corresponding volume value of the target to be measured is established, the number of the pixel points whose volume values are closer to each other is classified into one class, each class is distinguished according to the size of the volume value, and the preset deviation is used as a gradient. And fitting the pixel number and the volume value classification data trend by using an approximate normal distribution function, taking the pixel number in a normal distribution confidence interval as a total effective point, calculating all volume values in the effective point, and then calculating an average value to further obtain an estimated volume which is closer to the actual target volume to be measured.
In one application scenario, the preset deviation is between 3 and 6%.
According to an embodiment of the present invention, the step of obtaining the spectral information of the target to be measured and inputting the spectral information into a spectral density model to obtain the estimated density of the target to be measured specifically includes:
traversing the sample pool of the spectral density model according to the spectral information of the target to be detected to obtain sample spectral information corresponding to the spectral information of the target to be detected;
and acquiring a sample density corresponding to the sample spectrum information, and taking the sample density as an estimated density.
Specifically, this embodiment provides an implementation manner for obtaining the estimated density of the target to be measured according to the spectral density model, where spectral information of the target to be measured is input through the spectral density model trained in advance, matching is performed in the spectral density model, a sample spectrum corresponding to the spectrum of the target to be measured is obtained, and the sample density of the sample is retrieved.
According to an embodiment of the present invention, the step of training the spectral density model based on sample spectral information of a sample and sample density corresponding to the sample spectral information specifically includes:
placing the sample with known sample mass into a second container, and filling inert gas into the second container;
obtaining a first volume of the sample mixed with the inert gas and a first pressure within the second container at the first volume;
maintaining a constant temperature in the second container, changing the volume of the second container and obtaining a second volume of the second container, and a second pressure in the second container at the second volume;
calculating a sample volume of the sample according to the first volume, the second volume, the first pressure and the second pressure based on the Bowman's law;
obtaining the sample density according to the sample mass and the sample volume;
acquiring sample spectrum information of the sample, and training according to the sample spectrum information and the sample density to obtain a spectrum density sample;
and repeating the steps for X times to obtain X spectral density samples, and obtaining the spectral density model according to the X spectral density samples.
Specifically, the embodiment provides an implementation manner of training a spectral density model, in which a sample volume is obtained by using the pomma law according to a sample with known sample mass, so as to obtain a sample density of the sample, a spectrum of the sample is collected at the same time, a spectral density sample corresponding to the sample spectrum and the sample density is established, and the above steps are repeated for many times to train, so as to obtain the spectral density model.
It should be noted that the sample mass can be weighed in advance by a balance.
According to an embodiment of the present invention, the step of inputting the estimated quality and the spectral information of the target to be measured into the spectral nutrient content model of the target to be measured to obtain the nutrient content of the target to be measured specifically includes:
traversing the sample pool of the spectral nutrient content model according to the spectral information of the target to be detected to obtain sample spectral information corresponding to the spectral information of the target to be detected;
acquiring the nutrient content of the sample in unit mass corresponding to the spectral information of the sample and the share of the estimated mass in unit mass;
and obtaining the nutrient content of the target to be detected according to the unit mass share contained in the estimated mass and the sample nutrient content contained in each unit mass.
Specifically, this embodiment provides an implementation manner for obtaining the nutrient content of the target to be measured according to a spectral nutrient content model, where the spectral information of the target to be measured is used to obtain sample spectral information in the spectral nutrient content model, the nutrient content of the sample in unit mass of the sample spectral information is obtained, and the unit mass share included in the estimated mass is obtained according to the unit mass of the sample spectral information, so as to obtain the nutrient content of the target to be measured.
According to an embodiment of the present invention, the step of training the spectral nutrient content model based on sample spectral information of a sample and sample nutrient content corresponding to the sample spectral information specifically includes:
obtaining the sample nutrient content of the sample in unit mass;
acquiring sample spectrum information of the sample;
training according to the sample spectrum information of the sample and the sample nutrient content in unit mass to obtain a spectrum nutrient content sample;
and repeating the steps for Y times to obtain Y spectral nutrient content samples, and obtaining the spectral nutrient content model according to the Y spectral nutrient content samples.
Specifically, the embodiment provides an implementation manner of training a spectrum nutrient content model, a spectrum nutrient content sample is established according to the obtained nutrient content contained in a unit mass sample and sample spectrum information corresponding to the sample, and the steps are repeated for many times to perform training, so that the spectrum nutrient content model is obtained.
According to the second aspect of the invention, a system based on the above intelligent detection method for determining the nutrient content of the dish food is provided, which comprises: the device comprises a first shell, a second shell, an image detection module and a spectrum detection module;
the first shell is arranged above the second shell and can be buckled with the second shell to form an accommodating chamber for accommodating a target to be detected;
the image detection module is connected with the first shell and used for acquiring image information of the target to be detected;
the spectrum detection module is connected with the second shell and used for collecting spectrum information of the target to be detected;
wherein the first container is within the second housing.
According to one embodiment of the invention, at least the second housing is of cylindrical configuration.
Particularly, the embodiment provides an implementation manner of the second shell, and the second shell is set to be a cylindrical structure, so that the bottom radius of the second shell is convenient to obtain, and further, the estimation of the volume of the object to be measured is performed through the image of the object to be measured.
According to an embodiment of the invention, the second housing is an inverted conical cylinder structure.
One or more technical solutions in the present invention have at least one of the following technical effects: the method and the system for rapidly detecting the nutrient content in the dish food by the fusion map realize the identification of the type of the dish food, the establishment of the image and the volume model of the dish food, the establishment of the spectrum and the density model of the dish food and the establishment of the spectrum and the nutrient content model of the dish food by utilizing the optical map technology, the machine learning and the data modeling algorithm, and further comprehensively estimate the quality and the nutrient content of the dish food according to the volume of the dish.
Additional aspects and advantages of the invention will be set forth in part in the description which follows and, in part, will be obvious from the description, or may be learned by practice of the invention.
Drawings
In order to more clearly illustrate the technical solutions of the present invention or the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
FIG. 1 is a schematic flow chart of a method for rapidly detecting the nutrient content in a dish food by using a fusion map provided by the invention;
FIG. 2 is one of the schematic diagrams of the assembly relationship of the detection device for intelligently detecting the nutrient content of the dish and food provided by the invention;
FIG. 3 is a second schematic view of the assembly relationship of the detecting device for intelligently detecting the nutritional content of the dishes provided by the present invention;
FIG. 4 is one of schematic diagrams of arrangement relations of collected target image information to be detected in the method for rapidly detecting the nutrient content in dishes and foods by using the fusion map provided by the invention;
FIG. 5 is a second schematic diagram of the arrangement relationship of the collected target image information in the method for rapidly detecting the nutrient content in the dishes by using the fusion map according to the present invention;
fig. 6 is a schematic structural diagram of an electronic device provided in the present invention.
Reference numerals:
10. a first housing; 20. A second housing; 30. A first container;
40. an image detection module; 50. A bottom spectrum detection module; 60. A side spectrum detection module;
70. a target to be measured; 810: a processor; 820: a communication interface;
830: a memory; 840: a communication bus.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be obtained by a person skilled in the art without any inventive step based on the embodiments of the present invention, are within the scope of the present invention.
The present application will now be described in detail with reference to the drawings, and the specific operations in the method embodiments may also be applied to the apparatus embodiments or the system embodiments. In the description of the present application, "at least one" includes one or more unless otherwise specified. "plurality" means two or more. For example, at least one of A, B and C, comprising: a alone, B alone, a and B in combination, a and C in combination, B and C in combination, and A, B and C in combination. In this application, "/" means "or, for example, A/B may mean A or B; "and/or" herein is merely an association describing an associated object, and means that there may be three relationships, e.g., a and/or B, which may mean: a exists alone, A and B exist simultaneously, and B exists alone.
FIG. 1 is a flow chart of a method for rapidly detecting the nutrient content in dish food by using a fusion map provided by the invention. Fig. 1 shows a flow of the method for rapidly detecting the nutrient content in the dishes by using the fusion map provided by the invention, the invention obtains the image information of the object 70 to be detected, and obtains the estimated volume of the object 70 to be detected according to the image information.
Further, the spectral information of the target 70 to be measured is obtained, and the spectral information is input to the spectral density model to obtain the estimated density of the target 70 to be measured.
Further, the estimated mass of the target to be measured 70 is calculated according to the estimated volume and the estimated density, and the estimated mass and the spectrum information of the target to be measured 70 are input into the spectrum nutrient content model of the target to be measured 70 to obtain the nutrient content of the target to be measured 70.
It should be noted that, the object 70 to be measured mentioned in the present invention is a state after the dish food is mashed, and the internal state inside the object 70 to be measured is uniform and stable.
Fig. 2 and fig. 3 are one and two schematic diagrams of the assembly relationship of the detection device for intelligently determining the nutrient content of dishes provided by the invention. The detection device for intelligently determining the nutrient content of the dish and food is shown by the figures 2 and 3, the first container 30 is arranged in the second housing 20, and the spectrum detection module is arranged in the first container 30 and comprises a bottom spectrum detection module 50 and a side spectrum detection module 60.
It should be noted that the second container is not shown in the present invention, and the second container may be understood as a vessel for calculating the sample volume of the sample, and the sample density of the sample is directly obtained through the calculation of the sample volume and the advance acquisition of the sample mass.
Further, a sample of spectral density based on the sample is established by collecting sample spectral information of the sample, in combination with the sample density.
Fig. 4 and 5 are one and two schematic diagrams of arrangement relations of image information of a target 70 to be detected collected in the method for rapidly detecting the nutrient content in dishes and foods by using the fusion map provided by the invention. The process of image information acquisition of the object 70 to be measured is illustrated by fig. 4 and 5.
In an application scenario, a tomato egg dish is taken as a target to be detected, the protein content in the dish is taken as a nutrition parameter to be detected, as shown in fig. 4 and 5, after the target to be detected 70 is kept still for a period of time and the state of the minced dish food is stable, a light source of the image detection module 40 irradiates the target to be detected 70, a camera completes image identification of the target to be detected 70 and image acquisition of the boundary of the opening of the target to be detected 70 and the first container 30, and the image identification realizes identification of the target to be detected 70 by utilizing a machine learning algorithm, a convolutional neural network and the like; after the image is collected, taking a certain pixel point (i) of the object 70 to be measured on the boundary of the first container 30 as an example, the horizontal distance between the opening boundary of the object 70 to be measured and the horizontal projection of the containing area of the object 70 to be measured in the horizontal direction of the pixel point is obtained as j (i) f (i), wherein j (i) is a certain pixel point on the contour boundary of the object 70 to be measured, and f (i) is a projection point on the opening boundary of the first container 30. The first container 30 is in a circular truncated cone shape, the height is AE, the radius of the circular bottom surface with the larger upper bottom is R, the radius of the circular bottom surface with the smaller lower bottom is R, the projection of a certain pixel point (i) on the boundary of the object 70 to be measured in the vertical direction is j (i) d (i), namely f (i) E, the radius of the boundary point (i) on the upper surface of the first container 30 is s (i), the radius of the bottom surface of the lower surface is R, and the calculation formula of the estimated volume V of the object 70 to be measured in the first container 30 is as shown in formula (1):
wherein r is known, the value of S (i) can be obtained by processing the image of the object 70 to be measured, BJ (i) D (i) is obtained by using the projection size relation (2),
the estimated volume vsample (i) of the object 70 can be calculated by using the formulas (1), (2) and (3) based on the pixel point (i) of the object 70.
Fig. 6 is a schematic structural diagram of an electronic device provided in the present invention.
In some embodiments of the present invention, as shown in fig. 1, the present invention provides a method for rapidly detecting the nutrient content in a cooked food by using a fusion map, comprising:
the method comprises the following steps of (1) stirring dish food and filtering liquid to form a target to be detected 70, and placing the target to be detected 70 in a first container 30;
acquiring image information of the object 70 to be measured, and acquiring an estimated volume of the object 70 to be measured according to the image information;
acquiring spectral information of the target 70 to be measured, and inputting the spectral information into a spectral density model to obtain the estimated density of the target 70 to be measured, wherein the spectral density model is obtained by training based on sample spectral information of a sample and sample density corresponding to the sample spectral information;
and calculating to obtain the estimated mass of the target to be measured 70 according to the estimated volume and the estimated density, and inputting the estimated mass and the spectral information of the target to be measured 70 into a spectral nutrient content model of the target to be measured 70 to obtain the nutrient content of the target to be measured 70, wherein the spectral nutrient content model is obtained based on sample spectral information of the sample and sample nutrient content training corresponding to the sample spectral information.
In detail, the invention provides a method for rapidly detecting the nutrient content in dish food by fusing maps, which is used for solving the defect that the nutrient content of the dish food cannot be accurately detected in the prior art, and realizes the identification of the type of the dish food, the establishment of an image and dish food volume model, the establishment of a spectrum and dish food density model and the establishment of a spectrum and dish food nutrient content model by utilizing an optical map technology, a machine learning and data modeling algorithm, so that the quality and the nutrient content of the dish food are comprehensively estimated according to the volume of the dish.
In some possible embodiments of the present invention, as shown in fig. 4 and fig. 5, the step of obtaining image information of the object 70 to be measured, and obtaining the estimated volume of the object 70 to be measured according to the image information specifically includes:
collecting image information of the object 70 to be measured after standing in the first container 30;
acquiring a point at the edge of an image in the image information as a first pixel point, and calculating to obtain a first estimated volume of the target 70 to be measured according to the distance from the first pixel point to the edge of the first container 30, the distance from the first pixel point to the center of the image, the depth of the first container 30 and the bottom radius of the first container 30, wherein the first estimated volume is the volume of the target 70 to be measured based on the first pixel point;
repeating the steps until N estimated volumes of N pixel points are obtained, wherein N is more than or equal to two;
and calculating to obtain a first average value according to the N estimated volumes, and taking the first average value as the estimated volume.
Specifically, the embodiment provides an implementation manner for estimating the estimated volume of the object 70 to be measured according to the image information, the image of the object 70 to be measured is acquired by collecting the image of the object 70 to be measured, a pixel point on the object 70 to be measured in the image is extracted, the volume of the object 70 to be measured is estimated according to the pixel point, and after a plurality of estimated volumes are obtained by selecting a plurality of pixel points for measurement, an average value is obtained to obtain an estimated volume approximate to the object 70 to be measured.
In some possible embodiments of the present invention, the step of calculating a first average value according to the N estimated volumes and using the first average value as the estimated volume specifically includes:
acquiring a preset deviation, and establishing N normal distribution functions of the estimated volume by taking the preset deviation as a gradient interval;
and obtaining M estimated volumes within the normal distribution confidence interval, and calculating to obtain a first average value according to the M estimated volumes, wherein M is less than N.
Specifically, the present embodiment provides another implementation manner for estimating the estimated volume of the object to be measured 70 according to the image information, in order to obtain an estimated volume closer to the actual volume of the object to be measured 70, a functional relationship graph of each pixel point in the image and the corresponding volume value of the object to be measured 70 is established, the number of the pixel points whose volume values are closer is classified into one class, each class is distinguished according to the size of the volume value, and the preset deviation is used as a gradient. And fitting the pixel number and the volume value classification data trend by using an approximate normal distribution function, taking the pixel number in a normal distribution confidence interval as a total effective point, calculating all volume values in the effective point, and then calculating an average value to further obtain an estimated volume which is closer to the actual volume of the target 70 to be measured.
In one application scenario, the preset deviation is between 3 and 6%.
In some possible embodiments of the present invention, the step of obtaining the spectral information of the object 70 to be measured, and inputting the spectral information into the spectral density model to obtain the estimated density of the object 70 to be measured specifically includes:
traversing the sample pool of the spectral density model according to the spectral information of the target 70 to be measured, and obtaining sample spectral information corresponding to the spectral information of the target 70 to be measured;
and acquiring the sample density corresponding to the sample spectrum information, and taking the sample density as the estimated density.
Specifically, in the embodiment, an implementation manner of obtaining the estimated density of the target 70 to be measured according to the spectral density model is provided, and the spectral information of the target 70 to be measured is input through the spectral density model trained in advance, and matching is performed in the spectral density model to obtain a sample spectrum corresponding to the spectrum of the target 70 to be measured, and the sample density of the sample is retrieved.
In some possible embodiments of the present invention, the step of training the spectral density model based on sample spectral information of the sample and sample density corresponding to the sample spectral information specifically includes:
putting a sample with known sample mass into a second container, and filling inert gas into the second container;
obtaining a first volume of the sample mixed with the inert gas and a first pressure in the second container at the first volume;
maintaining a constant temperature in the second vessel, changing the volume of the second vessel and obtaining a second volume of the second vessel, and a second pressure in the second vessel at the second volume;
calculating to obtain a sample volume of the sample according to the first volume, the second volume, the first pressure and the second pressure based on the Boma's law;
obtaining the sample density according to the sample mass and the sample volume;
acquiring sample spectrum information of a sample, and training according to the sample spectrum information and sample density to obtain a spectral density sample;
and repeating the steps for X times to obtain X spectral density samples, and obtaining a spectral density model according to the X spectral density samples.
Specifically, the embodiment provides an implementation manner of training a spectral density model, in which a sample volume is obtained by using the pomma law according to a sample with known sample mass, so as to obtain a sample density of the sample, a spectrum of the sample is collected at the same time, a spectral density sample corresponding to the sample spectrum and the sample density is established, and the above steps are repeated for many times to perform training, so as to obtain the spectral density model.
It should be noted that the sample mass can be weighed in advance by a balance.
In some possible embodiments of the present invention, the step of inputting the estimated quality and the spectral information of the target to be measured 70 into the spectral nutrient content model of the target to be measured 70 to obtain the nutrient content of the target to be measured 70 specifically includes:
traversing a sample pool of the spectral nutrient content model according to the spectral information of the target 70 to be detected to obtain sample spectral information corresponding to the spectral information of the target 70 to be detected;
acquiring the nutrient content of the sample in unit mass corresponding to the spectral information of the sample, and estimating the unit mass share contained in the mass;
and obtaining the nutrient content of the target to be measured 70 according to the unit mass share contained in the estimated mass and the nutrient content of the sample contained in each unit mass.
Specifically, the embodiment provides an implementation manner for obtaining the nutrient content of the target to be measured 70 according to the spectral nutrient content model, the spectral information of the target to be measured 70 is used to obtain the spectral information of the sample in the spectral nutrient content model, the nutrient content of the sample in unit mass of the spectral information of the sample is obtained, and the unit mass share included in the estimated mass is obtained according to the unit mass of the spectral information of the sample, so that the nutrient content of the target to be measured 70 is obtained.
In some possible embodiments of the present invention, the step of training the spectral nutrient content model based on sample spectral information of the sample and sample nutrient content corresponding to the sample spectral information specifically includes:
obtaining the sample nutrient content of a sample in unit mass;
acquiring sample spectrum information of a sample;
training according to sample spectrum information of the sample and the sample nutrient content in unit mass to obtain a spectrum nutrient content sample;
and repeating the steps for Y times to obtain Y spectral nutrient content samples, and obtaining a spectral nutrient content model according to the Y spectral nutrient content samples.
Specifically, the embodiment provides an implementation manner of training a spectrum nutrient content model, a spectrum nutrient content sample is established according to the obtained nutrient content contained in a unit mass sample and sample spectrum information corresponding to the sample, and the steps are repeated for many times to perform training, so that the spectrum nutrient content model is obtained.
In some embodiments of the present invention, as shown in fig. 2 and fig. 3, the present scheme provides a system based on the above-mentioned intelligent detection method for determining the nutrient content of a dish food, comprising: a first housing 10, a second housing 20, an image detection module 40 and a spectrum detection module; the first casing 10 is disposed above the second casing 20, and can be fastened with the second casing 20 to form an accommodating chamber for accommodating the object 70 to be measured; the image detection module 40 is connected with the first casing 10 and is used for acquiring image information of the target 70 to be detected; the spectrum detection module is connected with the second shell 20 and used for collecting spectrum information of the target 70 to be detected; wherein the first container 30 is located within the second housing 20.
In detail, the invention also provides a detection system for intelligently determining the nutrient content of the dish food, which is used for solving the defect that the nutrient content of the dish food cannot be accurately detected in the prior art, and the identification of the type of the dish food, the establishment of an image and dish food volume model, the establishment of a spectrum and dish food density model and the establishment of a spectrum and dish food nutrient content model are realized by utilizing an optical spectrum technology, a machine learning and data modeling algorithm, so that the quality and the nutrient content of the dish food are comprehensively estimated according to the volume of the dish.
In some possible embodiments of the invention, at least the second housing 20 is of cylindrical configuration.
Specifically, the present embodiment provides an implementation manner of the second casing 20, and the second casing 20 is configured as a cylindrical structure, so as to facilitate obtaining the bottom radius of the second casing 20, and further enable estimating the volume of the object 70 to be measured through the image of the object 70 to be measured.
In some possible embodiments of the invention, the second housing 20 is an inverted conical cylinder structure.
Fig. 6 illustrates a physical structure diagram of an electronic device, which may include, as shown in fig. 6: a processor (processor)810, a communication Interface 820, a memory 830 and a communication bus 840, wherein the processor 810, the communication Interface 820 and the memory 830 communicate with each other via the communication bus 840. The processor 810 may call logic instructions in the memory 830 to perform the above-described method of fusing maps to rapidly detect the nutritional content in the dish food.
It should be noted that, when being implemented specifically, the electronic device in this embodiment may be a server, a PC, or other devices, as long as the structure includes the processor 810, the communication interface 820, the memory 830, and the communication bus 840 shown in fig. 6, where the processor 810, the communication interface 820, and the memory 830 complete mutual communication through the communication bus 840, and the processor 810 may call the logic instructions in the memory 830 to execute the above method. The embodiment does not limit the specific implementation form of the electronic device.
In addition, the logic instructions in the memory 830 may be implemented in software functional units and stored in a computer readable storage medium when the logic instructions are sold or used as independent products. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (10)
1. A method for rapidly detecting the nutrient content in dishes by fusing maps is characterized by comprising the following steps:
the method comprises the steps of stirring dish food and filtering liquid to form a target to be detected, and placing the target to be detected in a first container;
acquiring image information of the target to be detected, and acquiring the estimated volume of the target to be detected according to the image information;
acquiring spectral information of the target to be detected, and inputting the spectral information into a spectral density model to obtain the estimated density of the target to be detected, wherein the spectral density model is obtained by training based on sample spectral information of a sample and sample density corresponding to the sample spectral information;
calculating to obtain the estimated mass of the target to be detected according to the estimated volume and the estimated density, and inputting the estimated mass and the spectral information of the target to be detected into a spectral nutrient content model of the target to be detected to obtain the nutrient content of the target to be detected, wherein the spectral nutrient content model is obtained based on sample spectral information of a sample and sample nutrient content training corresponding to the sample spectral information.
2. The method for rapidly detecting the nutrient content in the dish food by fusing the maps as claimed in claim 1, wherein the step of obtaining the image information of the object to be detected and obtaining the estimated volume of the object to be detected according to the image information specifically comprises:
collecting image information of the target to be detected after the first container is stood;
acquiring a point at the edge of an image in image information as a first pixel point, and calculating to obtain a first estimated volume of the target to be measured according to the distance from the first pixel point to the edge of the first container, the distance from the first pixel point to the center of the image, the depth of the first container and the bottom radius of the first container, wherein the first estimated volume is the volume of the target to be measured based on the first pixel point;
repeating the steps until N estimated volumes of N pixel points are obtained, wherein N is more than or equal to two;
and calculating to obtain a first average value according to the N estimated volumes, and taking the first average value as the estimated volume.
3. The method for rapidly detecting the nutrient content in the dish food by using the fusion spectrum as claimed in claim 2, wherein the step of calculating a first average value according to the N estimated volumes and using the first average value as the estimated volume comprises:
acquiring a preset deviation and establishing N normal distribution functions of the pre-estimated volume by taking the preset deviation as a gradient interval;
and obtaining M estimated volumes within a normal distribution confidence interval, and calculating to obtain the first average value according to the M estimated volumes, wherein M is less than N.
4. The method for rapidly detecting the nutrient content in the dish food by fusing the maps as claimed in claim 1, wherein the step of obtaining the spectral information of the target to be detected and inputting the spectral information into a spectral density model to obtain the estimated density of the target to be detected specifically comprises:
traversing the sample pool of the spectral density model according to the spectral information of the target to be detected to obtain sample spectral information corresponding to the spectral information of the target to be detected;
and acquiring a sample density corresponding to the sample spectrum information, and taking the sample density as an estimated density.
5. The method for rapidly detecting the nutrient content in the dish food by fusing the maps as claimed in claim 4, wherein the spectral density model is trained based on sample spectral information of a sample and sample density corresponding to the sample spectral information, and specifically comprises:
placing the sample with known sample mass into a second container, and filling inert gas into the second container;
obtaining a first volume of the sample mixed with the inert gas and a first pressure within the second container at the first volume;
maintaining a constant temperature in the second container, changing the volume of the second container and obtaining a second volume of the second container, and a second pressure in the second container at the second volume;
calculating a sample volume of the sample according to the first volume, the second volume, the first pressure and the second pressure based on the Bowman's law;
obtaining the sample density according to the sample mass and the sample volume;
acquiring sample spectrum information of the sample, and training according to the sample spectrum information and the sample density to obtain a spectrum density sample;
and repeating the steps for X times to obtain X spectral density samples, and obtaining the spectral density model according to the X spectral density samples.
6. The method for rapidly detecting the nutrient content in the dish food by fusing the maps as claimed in claim 1, wherein the step of inputting the estimated quality and the spectral information of the target to be detected into the spectral nutrient content model of the target to be detected to obtain the nutrient content of the target to be detected specifically comprises:
traversing the sample pool of the spectral nutrient content model according to the spectral information of the target to be detected to obtain sample spectral information corresponding to the spectral information of the target to be detected;
acquiring the nutrient content of the sample in unit mass corresponding to the spectral information of the sample and the share of the estimated mass in unit mass;
and obtaining the nutrient content of the target to be detected according to the unit mass share contained in the estimated mass and the sample nutrient content contained in each unit mass.
7. The method for rapidly detecting the nutrient content in the dish food by fusing the maps as claimed in claim 6, wherein the spectral nutrient content model is trained based on sample spectral information of a sample and sample nutrient content corresponding to the sample spectral information, and specifically comprises:
obtaining the sample nutrient content of the sample in unit mass;
acquiring sample spectrum information of the sample;
training according to the sample spectrum information of the sample and the sample nutrient content in unit mass to obtain a spectrum nutrient content sample;
and repeating the steps for Y times to obtain Y spectral nutrient content samples, and obtaining the spectral nutrient content model according to the Y spectral nutrient content samples.
8. A system for intelligently measuring the detection method of the nutrient content of the dish food based on any one of the above claims 1 to 7 is characterized by comprising the following steps: the device comprises a first shell, a second shell, an image detection module and a spectrum detection module;
the first shell is arranged above the second shell and can be buckled with the second shell to form an accommodating chamber for accommodating a target to be detected;
the image detection module is connected with the first shell and used for acquiring image information of the target to be detected;
the spectrum detection module is connected with the second shell and used for collecting spectrum information of the target to be detected;
wherein the first container is within the second housing.
9. The intelligent detection device for determining the nutrient content of dish food as claimed in claim 8, wherein the second housing is a cylindrical structure.
10. The intelligent detection device for determining the nutritional content of cooked food as claimed in claim 8, wherein the second shell is an inverted cone-shaped cylinder structure.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110512979.0A CN113378882A (en) | 2021-05-11 | 2021-05-11 | Method and system for rapidly detecting nutrient content in dish food by fusing maps |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110512979.0A CN113378882A (en) | 2021-05-11 | 2021-05-11 | Method and system for rapidly detecting nutrient content in dish food by fusing maps |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113378882A true CN113378882A (en) | 2021-09-10 |
Family
ID=77572673
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110512979.0A Pending CN113378882A (en) | 2021-05-11 | 2021-05-11 | Method and system for rapidly detecting nutrient content in dish food by fusing maps |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113378882A (en) |
-
2021
- 2021-05-11 CN CN202110512979.0A patent/CN113378882A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105181912B (en) | A kind of Noninvasive Measuring Method of Freshness in rice storage | |
| US20120179665A1 (en) | Health monitoring system | |
| CN107658001B (en) | Household oil health management method and system | |
| CN104778374A (en) | Automatic dietary estimation device based on image processing and recognizing method | |
| CN109064509B (en) | Method, device and system for recognizing food volume and food heat | |
| CN104198695B (en) | A kind of method for the interpretation of result that developed the color to colloidal gold strip | |
| CN109959765B (en) | Salmon freshness detection system and salmon freshness detection method | |
| Akpro Hippocrate et al. | Food weight estimation using smartphone and cutlery | |
| CN108663367A (en) | A kind of egg quality lossless detection method based on egg unit weight | |
| JP2008204105A (en) | Automatic food intake measuring system and automatic dietary intake measuring method | |
| CN109431288A (en) | A kind of electric cooker control method, device, storage medium and electric cooker | |
| Akpa et al. | Smartphone-based food weight and calorie estimation method for effective food journaling | |
| CN113324928B (en) | Intelligent nondestructive detection method for nutrient content of dish food based on map data | |
| CN114566254A (en) | Non-contact computer vision intelligent diet nutrition assessment method, system and equipment | |
| CN109074861A (en) | Food monitors system | |
| CN113378882A (en) | Method and system for rapidly detecting nutrient content in dish food by fusing maps | |
| CN109781730A (en) | A kind of method of quick identification chalk rice | |
| JP7038335B1 (en) | Information provision method | |
| CN113874678A (en) | Food measuring method, device and program | |
| CN115187972A (en) | Dish identification method based on feature comparison | |
| An et al. | Viscocam: Smartphone-based drink viscosity control assistant for dysphagia patients | |
| Sadeq et al. | Smartphone-based calorie estimation from food image using distance information | |
| CN111414784B (en) | Method and device for determining water adding level, mobile terminal and storage medium | |
| CN114295611B (en) | Mutton mixing assumption detection method based on visual imaging in time sequence temperature change process | |
| CN117078955B (en) | Health management method based on image recognition |
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 |