CN117110287B - Edible ice ball manufacturing quality detection method - Google Patents
Edible ice ball manufacturing quality detection method Download PDFInfo
- Publication number
- CN117110287B CN117110287B CN202311377131.7A CN202311377131A CN117110287B CN 117110287 B CN117110287 B CN 117110287B CN 202311377131 A CN202311377131 A CN 202311377131A CN 117110287 B CN117110287 B CN 117110287B
- Authority
- CN
- China
- Prior art keywords
- model
- quality detection
- surface quality
- ice
- test
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 59
- 238000004519 manufacturing process Methods 0.000 title claims abstract description 19
- 238000012360 testing method Methods 0.000 claims abstract description 32
- 238000012549 training Methods 0.000 claims abstract description 22
- 238000000034 method Methods 0.000 claims abstract description 16
- 230000007246 mechanism Effects 0.000 claims description 33
- 238000011156 evaluation Methods 0.000 claims description 8
- 230000002950 deficient Effects 0.000 claims description 6
- 238000010276 construction Methods 0.000 claims description 4
- 238000005259 measurement Methods 0.000 claims description 4
- 230000000694 effects Effects 0.000 claims description 3
- 230000004913 activation Effects 0.000 claims description 2
- 238000004458 analytical method Methods 0.000 claims description 2
- 238000004364 calculation method Methods 0.000 claims description 2
- 230000006870 function Effects 0.000 claims description 2
- 238000010030 laminating Methods 0.000 claims description 2
- 210000002569 neuron Anatomy 0.000 claims description 2
- 238000011176 pooling Methods 0.000 claims description 2
- 238000012545 processing Methods 0.000 abstract description 7
- 230000010365 information processing Effects 0.000 description 6
- 230000008569 process Effects 0.000 description 2
- 229910000831 Steel Inorganic materials 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010790 dilution Methods 0.000 description 1
- 239000012895 dilution Substances 0.000 description 1
- 238000004090 dissolution Methods 0.000 description 1
- 230000008014 freezing Effects 0.000 description 1
- 238000007710 freezing Methods 0.000 description 1
- 238000002844 melting Methods 0.000 description 1
- 230000008018 melting Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000002035 prolonged effect Effects 0.000 description 1
- 238000003908 quality control method Methods 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 239000010959 steel Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01G—WEIGHING
- G01G17/00—Apparatus for or methods of weighing material of special form or property
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/0464—Convolutional networks [CNN, ConvNet]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Abstract
The utility model provides a method for detecting the manufacturing quality of edible ice balls, which is suitable for the field of food processing and comprises the steps of image acquisition, image label adding, training set and test set dividing, model building, model training, model testing, model application, weight detection and quality detection completion; the method for detecting the manufacturing quality of the edible ice ball is suitable for the field of food processing and has the advantages of high accuracy, wide applicability and the like.
Description
Technical Field
The utility model relates to a method for detecting the manufacturing quality of edible ice balls, which is suitable for the field of food processing.
Background
The spherical edible ice has a great demand in the catering market because it is not only beautiful in appearance but also can exert the effect of iced. In addition, compared with ice cubes in other shapes, the dissolution speed of the spherical edible ice is slower, the cup space can be fully filled, the effective freezing time is prolonged in the process of mixing wine, and the influence of excessive dilution on the taste is avoided. Thus being popular in the wine industry. Currently, there are two main ways of making spherical ice on the market. The ice ball is made by directly using a die, but the ice ball made by the method has bubbles, is opaque, has poor ornamental value and higher melting speed, can influence the taste of the drink, and is not suitable for high-grade places. The other is that a keyer manually engraves the transparent ice cubes into spheres, but the mode has the problems of insufficient roundness, low efficiency, high cost and the like, and mass production cannot be performed.
The quality of the current edible ice ball is not guaranteed, the manufacturing quality of the ice ball is guaranteed, on one hand, no crack is needed to be guaranteed on the appearance of the edible ice ball, and then the ornamental value and certain taste of the appearance are guaranteed, on the other hand, the ice ball is guaranteed to have certain weight, and further the quality of each ice ball is guaranteed to avoid the condition of being too light or too heavy in a certain range, so that a consistent eating experience can be provided for consumers, and the using amount of ice cubes can be accurately controlled in the process of regulating wine. Weight detection also helps to ensure quality control of the manufacturing process to meet sanitary and food safety requirements. Based on the above, the patent provides a method for detecting the manufacturing quality of the edible ice ball.
Disclosure of Invention
The utility model aims to provide a method for detecting the manufacturing quality of edible ice balls, which is applicable to the field of food processing and has the advantages of high accuracy, wide applicability and the like.
The aim of the utility model can be achieved by adopting the following technical scheme:
s101, image acquisition is carried out,
the image acquisition comprises the steps of acquiring appearance images of n ice hockey balls, and arranging the appearance images into an image set C {1,2, 3..n }, wherein n is more than or equal to 100, and the appearance images adopt an image acquisition device to acquire appearance images of different positions of each ice hockey ball;
s102, adding an image label,
the image label adding comprises the steps of analyzing the appearance image of each ice hockey and adding a label, wherein the analysis comprises the steps of evaluating the surface quality of the ice hockey by adopting a manual identification method, assigning an image label F according to an evaluation result, assigning a label F=1 if the evaluation result is qualified, assigning a label F=0 if the evaluation result is unqualified, and finally evaluating the appearance images of all the ice hockey and obtaining the label to obtain an appearance image set C2{ F=1, F= 2,F =3..F=n|F=0 OR 1};
s103, dividing the training set and the testing set,
dividing the training set and the test set, namely dividing the exterior image set C2 containing the labels into the training set and the test set according to the proportion of 3:1, respectively placing the training set and the test set in different folders, and storing each picture in a text file in the form of picture path plus label for reading by a model;
s104, constructing a model,
the construction model comprises constructing an improved AlexNet network puck surface quality detection model aiming at a surface image set C2 containing labels, wherein the improved AlexNet network puck surface quality detection model is totally divided into 8 layers, the improved AlexNet network puck surface quality detection model is formed by laminating a convolution layer, an index linear unit (ELU) layer, a pooling layer and a full-connection layer, the AlexNet network puck surface quality detection model adopts an index linear unit (ELU) activation function, the expression is shown in the formula (1),
（1）
wherein, the value of alpha is 0.3;
the construction model further comprises that the improved AlexNet network puck surface quality detection model is set to be 64 in a parameter of Batchsize, 0.95 in momentum, 0.0005 in omega attenuation rate and 0.001 in learning rate, and adopts Droput regularization to eliminate all neurons according to 40% probability;
s105, training the model,
the model training comprises the step of carrying out model training on the improved AlexNet network puck surface quality detection model by adopting the training set data;
s106, performing model test,
the model test comprises the step of adopting the test set data to carry out the model test on the improved AlexNet network puck surface quality detection model, wherein the test effect adopts F1-measurement as an evaluation index, the calculation formula of the F1-measurement is shown as a formula (2),
(2)
wherein P is the accuracy of the model test result, R is the recall rate of the model test result, TP is a real case, FP is a false positive case, and FN is a false negative case;
the model test further comprises that the F1-Measure index is larger than or equal to 0.7, the improved AlexNet network puck surface quality detection model is considered to pass the test, otherwise, the set parameters are modified to carry out model training and model test again until the set parameters meet the requirements, and finally the trained improved AlexNet network puck surface quality detection model is output;
s107, the model application,
the model application comprises the steps that each ice hockey adopts the trained improved AlexNet network ice hockey surface quality detection model to detect the surface quality, if the model detection result is qualified, the next step is carried out, and unqualified ice hockey is put into a defective product collection mechanism;
s108, detecting the weight of the steel plate,
the weight detection comprises the steps of carrying out weight detection on the ice ball, entering the next step, wherein the weight meets the requirement, namely, the ice ball is placed into a defective product collecting mechanism, the weight of which does not meet the requirement, the weight detection result of the ice ball is within a specified threshold value, the specified threshold value is [0.6w, w ], and the w is the theoretical calculated weight of the ice ball;
s109, finishing the quality detection,
the quality detection is completed by putting the ice hockey which meets the requirements into a good product collecting mechanism.
The beneficial effects of the utility model are as follows: the utility model provides a method for detecting the manufacturing quality of edible ice balls, which has the advantages of high accuracy, wide applicability and the like.
Drawings
Fig. 1: the utility model relates to a flow chart of a method for detecting the manufacturing quality of edible ice balls.
Detailed Description
Specific embodiments of the present utility model will be described in detail below with reference to the accompanying drawings; it should be understood that the particular embodiments presented herein are illustrative and explanatory only and are not restrictive of the utility model.
The following is a specific example of a method for detecting the manufacturing quality of edible ice hockey.
The embodiment is based on an utility model patent number CN209341644U of the company, and the utility model patent name is an automatic ice ball cutting device.
The automatic ice hockey ball cutting device is characterized in that: the ice-making machine comprises a control system, and a thickness detection mechanism, an ice-cube shaving mechanism, a ball shaving mechanism, a quality detection mechanism and an ice outlet which are sequentially arranged, wherein the control system is respectively connected with the thickness detection mechanism, the ice-cube shaving mechanism, the ball shaving mechanism and the quality detection mechanism.
The quality detection mechanism is established and implemented according to the edible ice ball manufacturing quality detection method, and specifically comprises an image acquisition mechanism, a weight detection mechanism, an information processing mechanism and an execution mechanism, wherein the information processing mechanism stores a trained improved AlexNet network ice ball surface quality detection model; the quality detection mechanism comprises the following specific implementation steps of the edible ice ball manufacturing quality detection method according to the patent:
(1) The image acquisition mechanism acquires the appearance image of the ice hockey generated by the automatic ice hockey chipping mechanism (CN 209341644U);
(2) The information processing mechanism detects the acquired appearance image acquisition result by using the trained improved AlexNet network ice ball surface quality detection model to acquire a label prediction result, if the label prediction result is 1, the label prediction result is failed, the information processing mechanism transmits the label prediction result to the execution mechanism to put the ice ball into the defective product collection mechanism, and if the label prediction result is 0, the label prediction result is failed, no processing is performed;
(3) The image acquisition mechanism is followed by a weight detection mechanism, the ice ball passes through the weight detection mechanism to obtain a weight detection result, the information processing structure does not perform any processing according to the weight detection result if the weight detection result is within the threshold value [0.6w, w ], and if the weight detection result is not within the threshold value, the information processing structure transmits information to the execution mechanism to place the ice ball into the defective product collection mechanism.
(4) Finally, finishing the quality detection mechanism, and enabling the ice ball meeting the requirements to enter an ice outlet;
to sum up, the embodiment of the utility model relates to an edible ice ball manufacturing quality detection method, which mainly depends on an utility model patent of the company, namely an automatic ice ball cutting device (CN 209341644U).
In the embodiment, the utility model discloses a method for detecting the manufacturing quality of an edible ice hockey ball, which comprises the steps of image acquisition, image label adding, training set and test set dividing, model building, model training, model testing, model application, weight detection and quality detection completion; the method for detecting the manufacturing quality of the edible ice ball is suitable for the field of food processing and has the advantages of high accuracy, wide applicability and the like.
The foregoing description of the preferred embodiments of the utility model is not intended to limit the utility model to the particular embodiments disclosed, but on the contrary, the intention is to cover all modifications, equivalents, and alternatives falling within the spirit and scope of the utility model.
Claims (1)
1. The method for detecting the manufacturing quality of the edible ice ball is characterized by comprising the following steps of:
1) Collecting images;
2) Adding an image tag;
3) Dividing a training set and a testing set;
4) Constructing a model;
5) Training a model;
6) Testing a model;
7) Model application;
8) Weight detection;
9) Finishing quality detection;
the image acquisition comprises the steps of acquiring appearance images of n ice hockey balls, and arranging the appearance images into an image set C {1,2, 3..n }, wherein n is more than or equal to 100, and the appearance images adopt an image acquisition device to acquire appearance images of different positions of each ice hockey ball;
the image label adding comprises the steps of analyzing the appearance image of each ice hockey and adding a label, wherein the analysis comprises the steps of evaluating the surface quality of the ice hockey by adopting a manual identification method, assigning an image label F according to an evaluation result, assigning a label F=1 if the evaluation result is qualified, assigning a label F=0 if the evaluation result is unqualified, and finally evaluating the appearance images of all the ice hockey and obtaining the label to obtain an appearance image set C2{ F=1, F= 2,F =3..F=n|F=0 OR 1};
dividing the training set and the test set, namely dividing the exterior image set C2 containing the labels into the training set and the test set according to the proportion of 3:1, respectively placing the training set and the test set in different folders, and storing each picture in a text file in the form of picture path plus label for reading by a model;
the construction model comprises constructing an improved AlexNet network puck surface quality detection model aiming at a surface image set C2 containing labels, wherein the improved AlexNet network puck surface quality detection model is totally divided into 8 layers, the improved AlexNet network puck surface quality detection model is formed by laminating a convolution layer, an index linear unit (ELU) layer, a pooling layer and a full-connection layer, the AlexNet network puck surface quality detection model adopts an index linear unit (ELU) activation function, the expression is shown in the formula (1),
（1）
wherein, the value of alpha is 0.3;
the construction model further comprises that the improved AlexNet network puck surface quality detection model is set to be 64 in a parameter of Batchsize, 0.95 in momentum, 0.0005 in omega attenuation rate and 0.001 in learning rate, and adopts Droput regularization to eliminate all neurons according to 40% probability;
the model training comprises the step of carrying out model training on the improved AlexNet network puck surface quality detection model by adopting the training set data;
the model test comprises the step of adopting the test set data to carry out the model test on the improved AlexNet network puck surface quality detection model, wherein the test effect adopts F1-measurement as an evaluation index, the calculation formula of the F1-measurement is shown as a formula (2),
(2)
wherein P is the accuracy of the model test result, R is the recall rate of the model test result, TP is a real case, FP is a false positive case, and FN is a false negative case;
the model test further comprises that the F1-Measure index is larger than or equal to 0.7, the improved AlexNet network puck surface quality detection model is considered to pass the test, otherwise, the set parameters are modified to carry out model training and model test again until the set parameters meet the requirements, and finally the trained improved AlexNet network puck surface quality detection model is output;
the model application comprises the steps that each ice hockey adopts the trained improved AlexNet network ice hockey surface quality detection model to detect the surface quality, if the model detection result is qualified, the next step is carried out, and unqualified ice hockey is put into a defective product collection mechanism;
the weight detection comprises the steps of carrying out weight detection on the ice ball, entering the next step, wherein the weight meets the requirement, namely, the ice ball is placed into a defective product collecting mechanism, the weight of which does not meet the requirement, the weight detection result of the ice ball is within a specified threshold value, the specified threshold value is [0.6w, w ], and the w is the theoretical calculated weight of the ice ball;
the quality detection is completed by putting the ice hockey which meets the requirements into a good product collecting mechanism.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311377131.7A CN117110287B (en) | 2023-10-24 | 2023-10-24 | Edible ice ball manufacturing quality detection method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311377131.7A CN117110287B (en) | 2023-10-24 | 2023-10-24 | Edible ice ball manufacturing quality detection method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117110287A CN117110287A (en) | 2023-11-24 |
| CN117110287B true CN117110287B (en) | 2024-01-16 |
Family
ID=88795116
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311377131.7A Active CN117110287B (en) | 2023-10-24 | 2023-10-24 | Edible ice ball manufacturing quality detection method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117110287B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102135905A (en) * | 2011-03-17 | 2011-07-27 | 清华大学 | User customization-based body matching system and method |
| CN109840497A (en) * | 2019-01-30 | 2019-06-04 | 华南理工大学 | A kind of pointer-type water meter reading detection method based on deep learning |
| CN110717895A (en) * | 2019-09-24 | 2020-01-21 | 南京理工大学 | No-reference image quality evaluation method based on confidence score |
| CN114782391A (en) * | 2022-04-29 | 2022-07-22 | 广州大学 | Method, system and device for constructing defect detection model of few-sample industrial image |
-
2023
- 2023-10-24 CN CN202311377131.7A patent/CN117110287B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102135905A (en) * | 2011-03-17 | 2011-07-27 | 清华大学 | User customization-based body matching system and method |
| CN109840497A (en) * | 2019-01-30 | 2019-06-04 | 华南理工大学 | A kind of pointer-type water meter reading detection method based on deep learning |
| CN110717895A (en) * | 2019-09-24 | 2020-01-21 | 南京理工大学 | No-reference image quality evaluation method based on confidence score |
| CN114782391A (en) * | 2022-04-29 | 2022-07-22 | 广州大学 | Method, system and device for constructing defect detection model of few-sample industrial image |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117110287A (en) | 2023-11-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10653934B2 (en) | Systems, devices, and methods employing the same for enhancing audience engagement in a competition or performance | |
| CN110197218A (en) | Thunderstorm gale grade forecast classification method based on multi-source convolutional neural networks | |
| CN106405682B (en) | A kind of precipitation predicting method and device | |
| CN109034224A (en) | Hyperspectral classification method based on double branching networks | |
| CN103699948A (en) | Sports fitness management system based on Internet of Things technology and implementation method | |
| CN102542492A (en) | System and method for evaluating effect of visual advertisement | |
| CN104517162A (en) | On-line hardness forecasting method of continuous annealing product by means of integrated learning | |
| CN117110287B (en) | Edible ice ball manufacturing quality detection method | |
| CN103994937A (en) | Method for grading crisped grass carp meat quality on basis of texture analyzer | |
| CN109003134A (en) | Intelligent advertisement delivery system and method based on recognition of face | |
| CN113435593B (en) | Refrigeration equipment frosting prediction method based on sensor time sequence data analysis | |
| CN108932712A (en) | A kind of rotor windings quality detecting system and method | |
| CN107658001A (en) | A kind of home-use oily health control method and system | |
| CN109740072A (en) | Hotel's sort method and system under OTA platform based on POI | |
| CN114936799A (en) | Risk identification method and system in cement fiberboard construction process | |
| CN104639610A (en) | Method and system for pushing information | |
| CN106326451B (en) | A kind of webpage heat transfer agent block decision method of view-based access control model feature extraction | |
| CN114529819A (en) | Household garbage image recognition method based on knowledge distillation learning | |
| CN105334224A (en) | Automatic quality detection cloud platform | |
| CN107909042A (en) | A kind of continuous gesture cutting recognition methods | |
| CN110942182A (en) | Method for establishing typhoon prediction model based on support vector regression | |
| CN114863489A (en) | Movable intelligent construction site auxiliary inspection method and system based on virtual reality | |
| CN108151241A (en) | Air conditioning control method and device | |
| CN113781435A (en) | Cigarette packet appearance defect detection method based on YOLOV5 network | |
| CN108470184A (en) | Food materials recognition methods, identification device and household appliance |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |