阅读说明：本技术 金枪鱼鱼肉鲜度的检测方法 (Method for detecting freshness of tuna meat ) 是由 韩伟 施佩影 何英 洪东杰 于 2019-09-24 设计创作，主要内容包括：本发明涉及水产技术领域,具体涉及一种金枪鱼鱼肉鲜度的检测方法。通过采用金枪鱼样本训练检测模型。然后通过检测模型来预测检测金枪鱼鱼肉的鲜度等级。该方法具有操作简单、便捷,检测速度快,成本低的有点,并且该检测方法降低了对检测人员专业知识的要求；具有更好的普适性。(The invention relates to the technical field of aquatic products, in particular to a method for detecting the fish flesh freshness of tunas. The detection model was trained by using tuna samples. And then predicting the freshness grade of the tuna flesh through a detection model. The method has the advantages of simple and convenient operation, high detection speed and low cost, and reduces the requirement on professional knowledge of detection personnel; has better universality.)

1. A method for detecting the fish freshness of tunas is characterized by comprising the following steps:

(1) selecting a large number of tuna samples under different storage conditions, specifically, selecting the tuna samples at different storage times and different storage temperatures, and taking the tuna samples as a data set;

(2) collecting image information in the data set, and artificially dividing the tuna into different primary grades;

collecting fluorescence data information of the tuna samples in the data set, collecting and calculating corresponding K values of the tuna samples, and carrying out secondary grading on the tuna samples according to the K values;

respectively measuring impedance values of the tuna samples under alternating currents with different frequencies by adopting a voltammetry method;

(3) firstly, establishing a primary identification model of the meat quality of the tuna; training a primary identification model of the tuna through image information;

(4) establishing a secondary grade detection model for the tuna flesh freshness grade by adopting a neural network, and training fluorescence data information, impedance value information and K value obtained by tuna samples in data sets of different flesh qualities after the primary identification model identifies;

(5) and detecting the freshness grade of the tuna meat by using the primary grade identification model and the secondary grade detection model obtained by training.

2. The method of claim 1, wherein the value K is used as output information of the model.

3. The method for detecting the freshness of tuna meat according to claim 1, wherein the impedance value is processed as input data; the treatment method comprises the following steps: impedance values of at least 10 frequencies are tested and their variance is mapped to the K value.

4. The method of claim 1, wherein the amount of the tuna sample is determined by comparing the success rate of the trained models.

5. The method for detecting the freshness of tuna fish meat according to claim 1, wherein the optimal detection conditions are determined by appropriately adjusting the optimal detection conditions of fluorescence data and impedance data by comparing the predicted accuracy of the training models.

6. The method for detecting the freshness of tuna meat according to claim 1, wherein the tuna belongs to the same species or the same produced tuna product.

7. The method for detecting the freshness of tuna meat according to claim 1, wherein the primary ranking is determined based on the increase of storage time of the tuna under the same conditions.

Technical Field

The invention relates to the technical field of aquatic products, in particular to a method for detecting the fish flesh freshness of tunas.

Background

The value of aquatic products mainly depends on the freshness of the aquatic products. Freshness is one of the most important indicators of fish product quality, and tuna is used as one of the most important indicators

The method for detecting the freshness of the fish body mainly comprises a sensory evaluation method, a microbiological method, a physical and chemical method and the like. Sensory evaluation has certain artificial factors, and TVB-N is ammonia and lower amines generated by decomposition of aquatic products under the action of bacterial colonies and enzymes and is generally used as an index of freshness of meat. The most common method for measuring the freshness of fish meat is to calculate the K value of a nucleotide compound by chemical analysis; this method is often destructive to the fish themselves and is also time consuming. The index detection methods are long in time consumption and high in operation requirement, and cannot meet the requirement of rapid detection. Many scholars at home and abroad have studied new detection methods for fish freshness, such as image analysis technology, near infrared spectroscopy measurement, electronic nose technology, surface fluorescence spectroscopy, immobilized enzyme reactors, NSPEs membrane electrode method, liquid chromatography and the like, and have achieved certain results, but the methods are relatively complex.

Disclosure of Invention

The invention aims to solve the problems and provides a method for detecting the fish flesh freshness of tunas.

In order to achieve the purpose, the invention adopts the following technical scheme:

a method for detecting the fish freshness of tunas comprises the following steps:

(1) selecting a large number of tuna samples under different storage conditions, specifically, selecting the tuna samples at different storage times and different storage temperatures, and taking the tuna samples as a data set;

(2) collecting image information in the data set, and artificially dividing the tuna into different primary grades;

collecting fluorescence data information of the tuna samples in the data set, collecting and calculating corresponding K values of the tuna samples, and carrying out secondary grading on the tuna samples according to the K values;

respectively measuring impedance values of the tuna samples under alternating currents with different frequencies by adopting a voltammetry method;

(3) firstly, establishing a primary identification model of the meat quality of the tuna; training a primary identification model of the tuna through image information;

(4) establishing a secondary grade detection model for the tuna flesh freshness grade by adopting a neural network, and training fluorescence data information, impedance value information and a K value obtained by tuna samples in data sets with different flesh quality grades;

(5) and detecting the freshness grade of the tuna meat by using the primary grade identification model and the secondary grade detection model obtained by training.

Preferably, the value of K is used as output information of the model.

Preferably, specifically, the impedance value is processed as input data; the treatment method comprises the following steps: impedance values of at least 10 frequencies are tested and their variance is mapped to the K value.

Preferably, the amount of tuna samples is determined by comparing the success rate of the trained models.

Preferably, the optimal detection conditions for the fluorescence data and the impedance data are appropriately adjusted by comparing the predicted accuracy of the training model, and the optimal detection conditions are determined.

Preferably, the tuna is a tuna product of the same breed or the same place of production.

Preferably, the primary grade is determined according to the increase of the storage time of the tuna under the same conditions.

Compared with the prior art, the invention has the beneficial effects that: the method is used for detecting the freshness of the tuna products, has the advantages of simple and convenient operation, high detection speed and low cost, and reduces the requirement on professional knowledge of detection personnel; has better universality. Meanwhile, the method greatly reduces the workload and improves the working efficiency.

Detailed Description

The technical solution of the present invention is further described below by means of specific examples.

The raw materials used in the examples of the present invention are those commonly used in the art, and the methods used in the examples are those conventional in the art, unless otherwise specified.