CN-122017121-A - Automatic detection method and system for determining fatty acid value of paddy

CN122017121ACN 122017121 ACN122017121 ACN 122017121ACN-122017121-A

Abstract

The application discloses an automatic detection method and system for determining fatty acid values of paddy, wherein the method comprises the steps of determining paddy to be detected, obtaining a paddy leaching liquid sample to be detected, constructing a standardized optical environment, keeping illumination intensity stable, generating a stable rotating field, determining identification information of the paddy leaching liquid sample to be detected and associating basic information, synchronously executing pulse dripping and reaction solution image acquisition during titration, executing edge detection and liquid level identification on the acquired images to extract liquid area images, converting liquid area image pixel data into Lab or HSV color space, establishing a background reference data set, calculating color difference values in real time, establishing a color difference value trend curve, determining a color difference value change rate peak value, determining a titration end point in combination with a self-adaptive threshold, determining a titration amount, and inversely calculating the fatty acid values according to the determined titration amount and the associated paddy leaching liquid sample basic information. The application improves the anti-interference capability and the accuracy of the measurement of the fatty acid value of the paddy.

Inventors

Qian Yadan
WEI ZHIHUA
Lv Juncang
WANG RUOLIN
WEI JUNYAN

Assignees

中储粮陕西质检中心有限公司

Dates

Publication Date: 20260512
Application Date: 20260122

Claims (10)

1. An automatic detection method for determining fatty acid value of paddy, comprising: The method comprises the steps of determining the rice to be measured, generating weighing and extracting instructions aiming at the rice to be measured, controlling a weighing and extracting device to execute weighing, liquid adding, oscillating and filtering operations of the rice to be measured, and generating a rice leaching solution to be measured; The method comprises the steps of establishing a standardized optical environment, feeding back a dynamic adjustment light source through a light sensor to stabilize illumination intensity, starting a magnetic stirrer to generate a stable swirling flow field aiming at a rice leaching liquid sample to be tested, determining identification information of the rice leaching liquid sample to be tested and associating basic information before starting titration; When the dripping is started, the pulse dripping of the standard titration solution and the collection of the reaction solution image are synchronously executed; converting the pixel data of the liquid area image into Lab or HSV color space, and establishing a background reference data set; The method comprises the steps of establishing a color difference value trend curve based on sliding window smoothing, monitoring and calculating the color difference value change rate, and determining a color difference value change rate peak value; Determining a titration end point according to the color difference value trend curve, the color difference value change rate peak value and the self-adaptive threshold value, and determining the titration amount; and according to the corresponding dropping quantity of the current measured paddy leaching liquid sample and the related basic information of the paddy leaching liquid sample, automatically inverting to calculate the fatty acid value.
2. An automatic detection method for fatty acid value measurement of paddy according to claim 1, further comprising: The real-time turbidity of the current reaction solution is quantified by additionally arranging a photoelectric sensor and monitoring the transmitted light intensity and the scattered light intensity of the reaction solution in the titration process in real time, and the pretreatment parameters of the acquired image and the color difference value calculation parameters converted into Lab or HSV color space are dynamically adjusted according to the quantified real-time turbidity of the reaction solution.
3. An automatic detection method for fatty acid value measurement of paddy according to claim 1, further comprising: When the dripping is started, the reaction dynamics self-adaptive dripping of the standard titration solution is selected to be executed instead of the pulse dripping of the standard titration solution, and the reaction dynamics self-adaptive dripping comprises the following steps: Setting multi-stage titration strategies, presetting corresponding titration speed and single titration volume by different stage titration strategies, setting a preset color difference value change rate range matched with each stage titration strategy, wherein the larger the value of the preset color difference value change rate range is, the slower the titration speed and the smaller the single titration volume in the corresponding matched stage titration strategy are; and acquiring the color difference value change rate fed back at the preamble moment, and determining the preset color difference value change rate range where the fed back color difference value change rate is positioned to adaptively match with the titration strategy at the subsequent moment stage so as to assist in executing the reaction dynamics self-adaptive dropwise adding of the standard titration solution.
4. An automatic detection method for fatty acid value measurement of paddy according to claim 1, further comprising: Defining a fixed position RIO and a dynamic RIO in the process of executing edge detection and liquid level identification on the acquired image to extract a liquid region image, and extracting a liquid region image corresponding to a plurality of RIO regions; The method comprises the steps of respectively calculating color difference values of liquid region image frames corresponding to each RIO region and a background reference data set, establishing a color difference value trend curve corresponding to each RIO region, determining a color difference value change rate peak value corresponding to each RIO region, judging a titration endpoint according to the color difference value trend curve corresponding to each RIO region, the color difference value change rate peak value and a self-adaptive threshold value, recording the titration endpoint correspondingly judged by each RIO region as a candidate endpoint, checking whether the titration endpoints corresponding to all ROI regions are gathered within an allowable volume error range, selecting the median of the candidate endpoints as a final endpoint if the titration endpoints are gathered within the allowable volume error range, otherwise, considering that mixing is uneven or interference is too large, and generating a prompt or giving out a warning for prolonging the mixing time.
5. An automatic detection method for fatty acid value measurement of paddy according to claim 2, further comprising: The method comprises the steps of calculating and obtaining a corresponding matching liquid area image pixel data conversion mode according to a background color mean value in a current background reference data set and the real-time turbidity of a reaction solution, wherein the liquid area image pixel data conversion mode comprises the steps of converting liquid area image pixel data into a Lab color difference space matched with a local mean value smaller than a first preset color mean value and turbidity smaller than the first preset turbidity, converting liquid area image pixel data into an HSV color difference space matched with a local mean value larger than a second preset color mean value and turbidity larger than the second preset turbidity, converting liquid area image pixel data into a comprehensive color difference space matched with the local mean value in a first preset color mean value range and the second preset color mean value range and converting liquid area image pixel data into the comprehensive color difference space matched with the turbidity in the first preset turbidity range and the second preset turbidity range, converting the liquid area image pixel data into the Lab color difference space and the HSV color difference space, respectively calculating key components of each color space, and performing weighted fusion to generate the comprehensive color difference value.
6. An automatic detection method for fatty acid value measurement of paddy according to claim 3, further comprising: predicting the color difference value of the extracted liquid region image frame and the background reference data set by utilizing a deep learning algorithm and combining a titration strategy; Calculating the absolute difference value between the color difference value change rates corresponding to adjacent titration points at the current moment, and judging whether the absolute difference value between the color difference value change rates is larger than a preset color difference value change rate threshold value or not; If the absolute difference value is larger than a preset absolute difference value threshold or the difference value is larger than a preset color difference value change rate threshold, starting a color difference value calculation checking flow, wherein the color difference value calculation checking flow comprises the steps of reselecting and extracting a liquid area image, switching and converting a Lab or HSV color space, adjusting the weight of each component in the converted Lab or HSV value, and re-weighting and calculating the color difference value.
7. An automatic detection method for fatty acid value measurement of paddy according to claim 1, further comprising: And selecting to synchronously execute pulse dripping of the standard titration solution and multi-view acquisition of the reaction solution image, clustering to obtain the corresponding dripping amount of the current paddy leaching solution sample after the corresponding determined judging end point of the reaction solution image acquired at each view, and eliminating the corresponding dripping amount of the discrete paddy leaching solution sample.
8. An automatic detection system for determining fatty acid value of rice, comprising: the system comprises a rice leaching liquid sample acquisition module, a rice leaching liquid sample acquisition module and a rice processing module, wherein the rice leaching liquid sample acquisition module is used for determining rice to be measured; The system comprises a paddy leaching liquid titration image acquisition module, a magnetic stirrer, a standard titration solution pulse dripping and reaction solution image acquisition module, a liquid area image acquisition module and a liquid area image acquisition module, wherein the paddy leaching liquid titration image acquisition module is used for constructing a standardized optical environment, feeding back a dynamic adjustment light source through a light sensor to stabilize illumination intensity; The paddy leaching liquid drop color difference extraction module is used for converting the image pixel data of the liquid area into Lab or HSV color space and establishing a background reference data set; the paddy leaching liquor titration feature extraction module is used for establishing a chromatic aberration value trend curve based on sliding window smoothing treatment; the paddy leaching liquid drop quantitative determination module is used for determining a titration end point according to a color difference value trend curve, a color difference value change rate peak value and an adaptive threshold value to determine a drop quantity; And the rice leaching liquor fatty acid value calculation module is used for automatically inverting and calculating the fatty acid value according to the corresponding dropping quantity of the current measured rice leaching liquor sample and the related rice leaching liquor sample basic information.
9. A computer readable storage medium, characterized in that the computer readable storage medium comprises a stored computer program, wherein the computer program, when run, controls a device in which the computer readable storage medium is located to perform the method according to any one of claims 1 to 7.
10. A computer device comprising a memory, a processor and a program stored and executable on the memory, which when executed by the processor performs the steps of the method according to any one of claims 1 to 7.

Description

Automatic detection method and system for determining fatty acid value of paddy Technical Field The application relates to the technical field of detection of substance components, in particular to an automatic detection method and system for determining fatty acid values of rice. Background The rice fatty acid value is a core index for judging the aging degree and the storage quality grade of the rice, and has important significance in the grain industry in terms of ensuring the quality safety of the grain, reasonably planning the storage and circulation of the grain and the like. Along with the development of technology, the technology for detecting the fatty acid value of rice is continuously advanced, and an accurate and efficient detection method plays an important role in promoting the healthy development of the grain industry. In the prior art, the detection of the fatty acid value of rice generally relies on acid-base titration of the rice extract. Conventional glass electrode potentiometric titration methods are automated in that a change in potential is detected by a glass electrode to determine the endpoint of the titration. And the optical titration method based on photoelectric colorimetry has relatively low cost, and the photoelectric sensor is used for detecting the color change of the indicator to determine the titration endpoint. However, conventional glass electrode potentiometric titration has significant drawbacks in that the electrode core is expensive and is highly contaminated with lipids and proteins in the rice extract, which results in a delayed electrode response and a shortened lifetime. The optical titration method based on photoelectric colorimetry is greatly influenced by sample matrixes in practical application, and due to the fact that background colors of paddy leaching solutions in different production places and storage ages are remarkably different and are often accompanied by turbidity of different degrees, the conventional low-cost sensor is difficult to distinguish color-changing signals of indicators from background interference of samples, serious misjudgment easily occurs when a fixed threshold is used for judging a titration end point, and detection equipment is difficult to achieve low cost and high accuracy of results, so that detection accuracy and repeatability are poor. Disclosure of Invention The application provides an automatic detection method and system for determining the fatty acid value of paddy, aiming at solving the technical problems that the existing low-cost automatic detection equipment is difficult to overcome the problems of poor detection accuracy and repeatability caused by delay or misjudgment of titration end point identification due to the background color difference and turbidity interference of paddy leaching liquor. In a first aspect, the present application provides an automatic detection method for fatty acid value measurement of rice, comprising: The method comprises the steps of determining the rice to be measured, generating weighing and extracting instructions aiming at the rice to be measured, controlling a weighing and extracting device to execute weighing, liquid adding, oscillating and filtering operations of the rice to be measured, and generating a rice leaching solution to be measured; The method comprises the steps of establishing a standardized optical environment, feeding back a dynamic adjustment light source through a light sensor to stabilize illumination intensity, starting a magnetic stirrer to generate a stable rotation flow field for a paddy leaching liquid sample to be detected, determining identification information of the paddy leaching liquid sample to be detected and associating basic information before titration is started, synchronously executing pulse dripping of a standard titration solution and collection of a reaction solution image when titration is started, and executing edge detection and liquid level identification on the collected image to extract a liquid area image; converting the pixel data of the liquid area image into Lab or HSV color space, and establishing a background reference data set; The method comprises the steps of establishing a color difference value trend curve based on sliding window smoothing, monitoring and calculating the color difference value change rate, and determining a color difference value change rate peak value; Determining a titration end point according to the color difference value trend curve, the color difference value change rate peak value and the self-adaptive threshold value, and determining the titration amount; and according to the corresponding dropping quantity of the current measured paddy leaching liquid sample and the related basic information of the paddy leaching liquid sample, automatically inverting to calculate the fatty acid value. By adopting the scheme, the paddy to be measured is determined, automatic sample weighing and extraction are carried