Meat detection data visualization and traceability information intelligent presentation method

Abstract

The invention discloses a meat detection data visualization and traceability information intelligent presentation method, and relates to the field of consumer information services. The understanding rate of the detection data is increased to 90% through three-layer visualization of'result-basis-suggestion '; the traceability card displays after code scanning reduces the query time to 2 seconds; multi-language adaptation covers international users. The method solves the problems of information abstraction, complex interaction and narrow coverage of an existing scheme, improves the right to know and credibility of consumers, and is suitable for consumer query services in supermarket retail, meat import and other scenes.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of consumer-end food information service, in particular to a meat detection data visualization and traceability information intelligent presentation method adapted to a mobile phone APP, which is used for intuitively displaying meat detection results and full-chain traceability information to consumers, and belongs to the innovative application of the food safety information dissemination and traceability technical field. BACKGROUND

[0002] Consumers' demand for meat safety awareness is increasing, but the existing traceability query scheme has the problems of "abstract information, complex interaction, and narrow coverage": the detection results are mostly presented in the form of original numerical values (such as "spectral similarity 95%") or simple "qualified / unqualified" labels, and consumers cannot understand the detection basis; the traceability information is displayed in the form of a text list, and it takes 3-5 steps to view completely, and the query time is more than 10 seconds; there is a lack of multi-language adaptation, which cannot meet the query needs of imported meat consumers (such as foreign users), and the user coverage is limited.

[0003] In addition, the existing scheme does not associate the detection results with the physical characteristics of meat (such as the appearance characteristics corresponding to the adulteration area and quality grade), and the consumer trust is insufficient. Therefore, an intelligent presentation method is needed that can convert professional detection data into popular information, simplify the interaction process, and cover multi-language users. SUMMARY

[0004] (1) Technical problem to be solved 1. The detection data presentation is abstract, and the consumer understanding rate is less than or equal to 30%, and the detection basis cannot be intuitively obtained; 2. The traceability information interaction is complex, the query path is long (3-5 steps), the time consumption is more than 10 seconds, and the convenience is insufficient; 3. There is a lack of multi-language adaptation, which cannot cover imported meat consumers, and the user range is limited.

[0005] (2) Technical solution The present application realizes consumer-friendly information display through the three-layer design of "detection data hierarchical visualization-traceability information scenario presentation-multi-language intelligent adaptation", and the specific implementation is as follows: 1. Detection data hierarchical visualization Design a "result-basis-suggestion" three-layer display logic to convert professional data into popular information: ★ Result layer: present with "green qualified / red unqualified" icons + meat-specific icons (such as cartoon icons of beef and pork), and obtain the core conclusion within 1 second; ★ Basis layer: Generate detection heat map (based on Grad-CAM technology), label detection focus area (such as spectral peak deviation of adulterated meat, image texture abnormal area), click heat map to display popular explanation (such as "this spectral feature matches duck meat, adulteration ratio 3%"). ★ Suggestion layer: Based on detection results, push consumption guidance (such as "quality A level, recommend eating within 3 days" "slight adulteration, not recommended for purchase"), match consumer use scenarios.

[0006] 2. Traceability information scenario presentation Based on the "main code-subcode" traceability system, design "scan code to display" traceability card: ★ Main card: Display core information (detection time, detection agency, qualified conclusion) within 1 second after scanning the code, loading speed ≤1 second; ★ Expanded card: Click the main card to display "slaughter-transportation-retail" whole chain information, each link is associated with physical characteristics (such as slaughterhouse NFC tag information, transportation link temperature and humidity curve, retail terminal sampling record), subcode corresponds to specific parts of segmented meat (such as "beef tenderloin-number S001"); ★ Interaction optimization: Support gesture zoom (view details), voice reading (adapt to visually impaired users), operation steps ≤1 step.

[0007] 3. Multi-language intelligent adaptation Build Chinese, English, Japanese and Korean language packs to realize automatic switching and term unification: ★ Language switching: Automatically adapt based on the phone system language, support manual switching (set interface selection); ★ Term unification: Establish a detection-traceability term correspondence table (such as "adulteration" corresponds to "adulteration", "quality grading" corresponds to "quality grade"), ensure information accuracy; ★ Cultural adaptation: Adjust icon style for different language users (such as simple style for English interface, cartoon style for Japanese interface). BRIEF DESCRIPTION OF DRAWINGS

[0008] Figure 1 : Detection data "result-basis-suggestion" hierarchical visualization logic diagram Figure 2 : Traceability information scan presentation interaction process and card design diagram Figure 3 : Multi-language intelligent adaptation switching logic and interface schematic diagram (3) Beneficial effects 1. Understanding rate improved: Hierarchical visualization improves the understanding rate of consumers from 30% to 90%, solving the "information abstraction" pain point; 2. Query efficiency improvement: Traceability information "scan code to show", query time from 10 seconds to 2 seconds, operation steps simplified to 1 step, convenience significantly improved; 3. User coverage expansion: Multi-language adaptation covers imported meat consumers, user range expands by 30%, and consumer trust improves to 90%.

Claims

1. A method of meat inspection data visualization, characterized in that, Comprise: (1) Result layer: presented with "green pass / red fail" icon + meat-specific logo, response time ≤1s; (2) Basis layer: generate detection heat map (based on Grad-CAM technology), mark spectral peak shift and image texture abnormal area, click heat map to display popular explanation text (such as "this feature matches duck meat, with 3% adulteration"); (3) Suggestion layer: push consumption guidance (consumption period, risk warning) based on detection results, adapt to consumer usage scenarios.

2. The method of claim 1, wherein, The detection heat map is generated based on the feature weights output by the multi-modal model (spectrum + image), which can visually display the detection focus area.

3. A method for intelligent presentation of provenance information, characterized in that, Comprise: (1) Main card: display detection time, agency, and qualified conclusion within 1s after scanning, loading speed ≤1s; (2) Expanded card: click to display "slaughter-transport-retail" whole-chain information, each link is associated with physical characteristics (NFC tag, temperature and humidity curve), and sub-codes correspond to specific parts of segmented meat (such as "beef tenderloin-S001"); (3) Interaction optimization: support gesture zoom and voice reading, operation steps ≤1 step.

4. The method of claim 3, wherein, The traceability information is stored in the blockchain (Hyperledger Fabric), and the data is tamper-proof, ensuring the authenticity of the information.

5. A multi-lingual intelligent adaptation method, characterized by, Applied to the method of claims 1-4, comprising: (1) Integrate Chinese, English, Japanese, and Korean language packs, automatically switch based on the phone system language, and support manual selection; (2) Establish a detection-traceability terminology comparison table, with a consistency of ≥98%; (3) Adjust icon style for different language users to adapt to cultural habits.

6. A mobile phone APP, characterized in that, Integrate the methods of claims 1-5, support detection data hierarchical visualization, traceability information scanning presentation, and multi-language switching, and adapt to consumer's convenient inquiry needs.

Images