AI 牛只评估工具开发

4.0

15/06/2025

pt农业技术AI牛只评估图像识别农业技术PDF报告

Prompt

create a program in Portuguese to use the AI API to read the images generated with the camera and photo. Follow the information below
Objective:

To create an AI-driven tool that automates the process of cattle assessment. The tool will capture an image of an animal, analyze it to determine its weight (in kilograms and arrobas), identify its category (Ox, Calf, Bull, or Heifer) and color, and then generate a standardized PDF report with these findings.

AI System Persona and Capabilities:
You are an expert AI agricultural analyst specializing in computer vision for livestock management. You are proficient in:

Image Recognition and Analysis: Identifying and isolating cattle within an image, even in complex farm environments.
Biometric Estimation: Accurately estimating the live weight of cattle from a 2D image by analyzing body volume, skeletal structure, and other visual cues. You have been trained on a vast dataset of cattle images with corresponding weights.
Cattle Classification: Differentiating between various categories of cattle (Ox, Calf, Bull, Heifer) based on visual characteristics such as body shape, muscle definition, and reproductive organs.
Color Identification: Accurately determining and describing the color and pattern of the animal's coat.
Data Conversion: Converting weight from kilograms (kg) to arrobas (@), where 1 arroba = 15 kg.
PDF Report Generation: Compiling the analyzed data into a clear, concise, and professionally formatted PDF document.
User Interaction and Input:
The user will interact with the system through a simple form. The primary input will be an image of a single bovine animal, captured by a smartphone or other camera.

Form Fields:

Image Upload: A button to upload or capture a new image of the animal.
[Optional] Animal ID/Tag: A text field for the user to manually enter an ear tag number or any other unique identifier for the animal.
[Optional] Date of Reading: A date picker that defaults to the current date.
[Optional] Farm/Pasture Name: A text field to input the location of the reading.
Generate Report Button: A button that initiates the AI analysis and PDF generation.
AI Analysis and Processing Steps:
Upon receiving the image, the AI will perform the following actions:

Image Pre-processing:

Validate that the uploaded file is an image.
Isolate the primary bovine subject from the background.
Assess image quality (e.g., clarity, lighting, animal posture) and notify the user if the image is unsuitable for accurate analysis, providing brief guidance on capturing a better picture (e.g., "Please provide a clear side-view image of the animal standing on level ground.").
Weight Estimation:

Analyze the animal's body volume, length, and height from the image.
Apply a deep learning regression model to estimate the live weight in kilograms. The model should be trained on a diverse dataset of cattle breeds common in Brazil (e.g., Nelore, Guzerá, Brahman).
Convert the estimated weight in kilograms to arrobas using the formula:
Weight in Arrobas=
15
Weight in kg

The result should be displayed to two decimal places.
Cattle Classification:

Analyze the animal's morphology to classify it into one of the following categories:
Bull: Identify prominent masculine features, including a muscular neck and shoulders (with a noticeable crest or hump), a broader facial structure, and the presence of a scrotal sac.
Ox (Steer): Identify as a castrated male. The animal will lack the prominent muscular development of a bull and will not have testes. The body shape will be more blocky than a cow's.
Heifer: Identify as a young female that has not yet calved. The animal will have a more feminine appearance than a bull or ox, with less defined musculature and visible teats but no developed udder.
Calf: Identify as a very young animal of either sex, characterized by its smaller size and less developed features relative to a mature animal.
Color Identification:

Analyze the coat of the animal and describe its color using standard terms (e.g., Black, White, Red, Brown, Grey, Spotted [Black and White, Red and White], Brindle).
Output: PDF Report
The AI will generate a single-page PDF report with the following structure and content:

Header:

Report Title: "Relatório de Leitura de Gado" (Cattle Reading Report)
Farm/Pasture Name: (From optional user input)
Date of Reading: (From optional user input or current date)
Animal Information Section:

Image of the Animal: A cropped and centered image of the analyzed bovine.
Animal ID/Tag: (From optional user input)
AI Analysis Results:
Categoria (Category): [e.g., Novilha (Heifer)]
Cor (Color): [e.g., Branco (White)]
Peso Estimado (Estimated Weight):
[Weight in kg] kg
[Weight in @] @
Footer:

Generated by: "Gerado por Inteligência Artificial da [Your Company/System Name]" (Generated by the Artificial Intelligence of [Your Company/System Name])
Disclaimer: "O peso e a classificação são estimados com base na análise de imagem e podem variar. Para dados precisos, utilize métodos de pesagem e avaliação zootécnica tradicionais." (The weight and classification are estimated based on image analysis and may vary. For precise data, use traditional weighing and zootechnical evaluation methods.)
Example Workflow:
A Brazilian farmer wants to assess a heifer in the pasture.
He opens the application on his smartphone.
He enters the heifer's ear tag number: "BR012345".
He takes a side-view picture of the heifer using the app's camera function.
He clicks "Gerar Relatório" (Generate Report).
The AI analyzes the image. It estimates the heifer's weight to be 360 kg. It classifies the animal as a "Novilha" (Heifer) and identifies its color as "Branco" (White). It calculates the weight in arrobas as 24 @ (360 / 15).
A PDF file is generated and displayed on the farmer's phone, which he can then save or share. The PDF will contain the heifer's picture, her ID, and the analyzed data neatly formatted.
This detailed prompt provides a clear blueprint for developing an AI-powered tool that is both technologically advanced and practically useful for the target user base in the Brazilian cattle industry.

Resultado Gerado

import React, { useState } from 'react';

const App = () => {
 const [image, setImage] = useState(null);
 const [report, setReport] = useState(null);

 const handleImageUpload = (event) => {
 setImage(URL.createObjectURL(event.target.files[0]));
 };

 const generateReport = () => {
 // Simulate AI analysis
 const mockReport = {
 category: 'Novilha',
 color: 'Branco',
 weightKg: 360,
 weightArrobas: 24,
 };
 setReport(mockReport);
 };

 return (
 <div className="min-h-screen bg-gray-100 p-8">
 <div className="max-w-4xl mx-auto bg-white rounded-lg shadow-lg p-6">
 <h1 className="text-3xl font-bold text-gray-800 mb-6">Relatório de Leitura de Gado</h1>
 <div className="space-y-6">
 <div>
 <label className="block text-sm font-medium text-gray-700">Carregar Imagem</label>
 <input type="file" onChange={handleImageUpload} className="mt-1 block w-full" />
 </div>
 {image && (
 <div className="flex justify-center">
 <img src={image} alt="Animal" className="rounded-lg shadow-md" />
 </div>
 )}
 <button onClick={generateReport} className="w-full bg-blue-500 text-white py-2 px-4 rounded-md hover:bg-blue-600">Gerar Relatório</button>
 {report && (
 <div className="mt-6 space-y-4">
 <h2 className="text-xl font-semibold text-gray-800">Resultados da Análise</h2>
 <div className="grid grid-cols-2 gap-4">
 <div className="bg-gray-50 p-4 rounded-lg">
 <p className="text-sm text-gray-600">Categoria</p>
 <p className="text-lg font-medium text-gray-800">{report.category}</p>
 </div>
 <div className="bg-gray-50 p-4 rounded-lg">
 <p className="text-sm text-gray-600">Cor</p>
 <p className="text-lg font-medium text-gray-800">{report.color}</p>
 </div>
 <div className="bg-gray-50 p-4 rounded-lg">
 <p className="text-sm text-gray-600">Peso Estimado</p>
 <p className="text-lg font-medium text-gray-800">{report.weightKg} kg</p>
 </div>
 <div className="bg-gray-50 p-4 rounded-lg">
 <p className="text-sm text-gray-600">Peso em Arrobas</p>
 <p className="text-lg font-medium text-gray-800">{report.weightArrobas} @</p>
 </div>
 </div>
 </div>
 )}
 </div>
 </div>
 <div className="mt-8 text-center text-sm text-gray-500">
 Gerado por Inteligência Artificial da CattleAI. © 2025
 </div>
 </div>
 );
};

export default App;

Open Sandbox

Mais Recomendações

Ver Todos

每日牛奶记录App

农业技术en

4.0

Create a daily cow milk record app

牛奶记录农业管理App开发30/03/2025