Aqbeż għall-kontenut ewlieni
Malti— Agħżel lingwa
Menu

Qabbel l-Immaġni bl-API tal-Viżjoni

Lezzjoni 10 / 12 Aġġornat 2 ta’ Novembru 2026 Zviluppatur
F'din il-paġna
  1. Analiżi Komparattiva Viżiva
  2. Paragun Qabel/Wara
  3. Kontroll tal-Kwalità tal-Prodott
  4. Traċċar tal-Evoluzzjoni Maż-Żmien
  5. Paragun tal-Interfaċċji tal-Utent
  6. L-Aħjar Prattiki għall-Paragun
  7. Punti Ewlenin

Analiżi Komparattiva Viżiva

Il-paragun tal-immaġni huwa każ ta’ użu qawwi li jisfrutta l-kapaċità multi-immaġni koperta fil-Lezzjoni 6. Il-mistoqsija tinbidel minn “X’turi din l-immaġni?” għal “X’inbidel?”—operazzjoni li mudell semantiku jimmanigġja b’mod fundamentali differenti minn komparatur tal-pixels. Fejn differenza fil-pixels tissuġġerixxi diverġenza mingħajr ma tikkwalifikaha, il-mudell jista’ jiddikjara li tikketta ċċaqlaq b’żewġ millimetri u li logo għabet, imbagħad jiġġudika liema miż-żewġ affarijiet hija importanti.

Kontroll tal-kwalità, traċċar tal-evoluzzjoni, detezzjoni tar-regressjoni viżiva: din il-lezzjoni turi kif tistruttura prompts ta’ paragun u tisfrutta r-riżultati.

Paragun Qabel/Wara

L-aktar xenarju komuni jinvolvi l-verifika ta’ x’inbidel bejn żewġ stati. Il-messaġġ tas-sistema jiddefinixxi l-metodu ta’ paragun, filwaqt li l-messaġġ tal-utent jidentifika l-immaġni:

from mistralai.client import Mistral
import os

client = Mistral(api_key=os.environ["MISTRAL_API_KEY"])

response = client.chat.complete(
    model="mistral-large-latest",
    messages=[
        {
            "role": "system",
            "content": """You are an expert in comparative image analysis. When presented with two images:
1. Briefly describe each image
2. List identified differences (with location)
3. Assess the significance of each difference (minor/major)
4. Provide an overall similarity score (0-100%)"""
        },
        {
            "role": "user",
            "content": [
                {"type": "text", "text": "BEFORE Image (reference):"},
                {"type": "image_url", "image_url": "https://exemple.com/avant.jpg"},
                {"type": "text", "text": "AFTER Image (current state):"},
                {"type": "image_url", "image_url": "https://exemple.com/apres.jpg"},
                {"type": "text", "text": "Identify all differences between these two images."}
            ]
        }
    ]
)

print(response.choices[0].message.content)

L-interkalazzjoni ta’ test (“BEFORE Image”, “AFTER Image”) tgħin lill-mudell jidentifika r-rwol ta’ kull immaġni. It-tielet pass fil-prompt tas-sistema huwa dak li jagħmel ir-rapport azzjonabbli: mingħajr prijoritizzazzjoni, tirċievi lista fejn bidla fid-dell tiżen daqs parti nieqsa.

Kontroll tal-Kwalità tal-Prodott

Fl-aħħar tal-pipeline, dan ir-rapport irid jinqara minn programm, mhux minn bniedem. Għalhekk, ngħaqqdu l-paragun mal-modalità JSON mil-Lezzjoni 9: il-mudell jirritorna verdett, punteġġ, u lista ta’ difetti kkwalifikati li l-kodiċi jista’ jisfretta immedjatament.

import base64
import json

def encode_image(path):
    with open(path, "rb") as f:
        b64 = base64.standard_b64encode(f.read()).decode("utf-8")
    return f"data:image/jpeg;base64,{b64}"

def quality_control(reference_image, sample_image):
    """Compares a sample to its reference and detects defects."""
    response = client.chat.complete(
        model="mistral-large-latest",
        response_format={"type": "json_object"},
        messages=[
            {
                "role": "system",
                "content": "You are a quality inspector. Compare the sample to the reference and return a JSON report."
            },
            {
                "role": "user",
                "content": [
                    {"type": "text", "text": "Reference (compliant product):"},
                    {"type": "image_url", "image_url": encode_image(reference_image)},
                    {"type": "text", "text": "Sample to inspect:"},
                    {"type": "image_url", "image_url": encode_image(sample_image)},
                    {
                        "type": "text",
                        "text": """Compare and return:
{
  "compliant": true/false,
  "compliance_score": 0-100,
  "defects": [
    {"description": "...", "severity": "minor/major/critical", "location": "..."}
  ],
  "recommendation": "accept/review/reject"
}"""
                    }
                ]
            }
        ]
    )

    return json.loads(response.choices[0].message.content)

# Usage
result = quality_control("reference/product-ok.jpg", "batch-42/sample-7.jpg")

if result["compliant"]:
    print(f"COMPLIANT - Score: {result['compliance_score']}%")
else:
    print(f"NON-COMPLIANT - {len(result['defects'])} defect(s) detected")
    for defect in result["defects"]:
        print(f"  - [{defect['severity']}] {defect['description']}")

Il-kamp recommendation huwa aktar interessanti milli jidher: billi tiddistingwi bejn “review” u “reject”, tirriżerva l-intervent uman għal każijiet ambigwi minflok ma timponieh kullimkien jew l-ebda mkien.

Traċċar tal-Evoluzzjoni Maż-Żmien

Serje temporali—sit ta’ kostruzzjoni, tkabbir tal-għelejjel, degradazzjoni tal-istruttura—tiġi ttrattata bħala paragun estiż, sakemm timmarka kull immaġni bid-data tagħha. Il-mudell ikollu mbagħad l-assi temporali u jista’ jikkwalifika ritmu, mhux biss lakuna.

def analyze_evolution(chronological_images, context=""):
    """Analyzes visible evolution across a series of ordered images."""
    content = [
        {
            "type": "text",
            "text": f"Here are {len(chronological_images)} images taken at different times, in chronological order. {context}"
        }
    ]

    for i, (date, path) in enumerate(chronological_images):
        content.append({"type": "text", "text": f"Image {i+1} ({date}):"})
        content.append({"type": "image_url", "image_url": encode_image(path)})

    content.append({
        "type": "text",
        "text": "Describe the evolution observed between each step. Identify major changes and the general trend."
    })

    response = client.chat.complete(
        model="mistral-large-latest",
        messages=[{"role": "user", "content": content}]
    )

    return response.choices[0].message.content

# Construction site tracking
images = [
    ("2025-01", "site/jan-2025.jpg"),
    ("2025-06", "site/jun-2025.jpg"),
    ("2026-01", "site/jan-2026.jpg"),
]

evolution = analyze_evolution(images, context="Tracking a construction site.")
print(evolution)

Il-mistoqsija finali tiżdied biss wara l-immaġni kollha, ladarba l-fajl sħiħ ikun imqiegħed.

Paragun tal-Interfaċċji tal-Utent

L-iżvilupp tal-web joffri każ ta’ użu immedjat: il-konfrontazzjoni tal-mockup approvat mar-rendering attwalment prodott. Fejn test ta’ regressjoni tal-pixels ifalli fuq ċaqliqa ta’ pixel waħda mingħajr importanza, il-mudell jikkwalifika l-impatt ta’ kull diskrepanza.

def compare_interfaces(expected_screenshot, actual_screenshot):
    """Compares two interface screenshots to detect visual regressions."""
    response = client.chat.complete(
        model="mistral-large-latest",
        response_format={"type": "json_object"},
        messages=[
            {
                "role": "system",
                "content": "You are a QA tester specializing in visual testing of web interfaces."
            },
            {
                "role": "user",
                "content": [
                    {"type": "text", "text": "Expected design (mockup):"},
                    {"type": "image_url", "image_url": encode_image(expected_screenshot)},
                    {"type": "text", "text": "Current render (screenshot):"},
                    {"type": "image_url", "image_url": encode_image(actual_screenshot)},
                    {
                        "type": "text",
                        "text": """Compare and return:
{
  "pixel_perfect": true/false,
  "differences": [
    {"element": "...", "expected": "...", "actual": "...", "impact": "minor/moderate/major"}
  ],
  "fidelity_score": 0-100,
  "verdict": "compliant/minor_adjustments/regression"
}"""
                    }
                ]
            }
        ]
    )

    return json.loads(response.choices[0].message.content)

Meta tinqabad mal-pipeline ta’ integrazzjoni kontinwa tiegħek, din iċ-ċekka timblokka l-konsenja biss fuq verdett ta’ regression.

L-Aħjar Prattiki għall-Paragun

Dawn l-eżempji jaqsmu ftit prinċipji li jiswew li jsiru espliċiti. Dejjem semmi l-immaġni tiegħek b’tikketta interkalata—“Reference”, “Sample”, “Before”, “After”—għax mingħajrha, il-mudell jassenja rwoli bbażati fuq każwalità fl-ordni. Żgura l-komparabbiltà tal-isparar: f’angoli u dawl differenti, il-mudell jissinjala diskrepanzi li ġejjin biss mill-kundizzjonijiet tal-foto. Uża Mistral Large 3, għax il-paragun jimobilizza raġunar fin li mudelli iżgħar mhux dejjem jirnexxielhom jimmasterjaw. Illimita ruħek għal żewġ jew tliet immaġni għal kull sejħa; lil hinn minn hekk, il-preċiżjoni komparattiva ddgħajjef. U sistematikament itlob format strutturat, b’punteġġ, differenzi elenkati, u verdett, jekk ir-riżultat irid jiġi pproċessat awtomatikament.

Punti Ewlenin

  • Il-paragun tal-immaġni jisfrutta l-kapaċità multi-immaġni tal-API
  • Interkala test spjegattiv bejn l-immaġni biex tiċċara r-rwoli tagħhom
  • Il-modalità JSON (response_format) hija ideali għal rapporti ta’ paragun awtomatizzati
  • Każijiet ta’ użu primarji: kontroll tal-kwalità, traċċar tas-siti, QA viżiv, qabel/wara
  • Dejjem uża Mistral Large 3 għal kompiti ta’ paragun li jeħtieġu ħafna riżorsi