AI learns to spot tomato diseases using nearly 9,000 field images

Phys · 🇱🇰 public Sri Lanka · Aug 17, 2026 · 675 words · By Charles Darwin University

Agricultural Innovation Sustainable Agriculture Artificial Intelligence in Farming

headphones Listen to the eFinder podcast briefing

Generate a natural audio summary of this story

Daily briefing

What to know about Agricultural Innovation

Researchers from Charles Darwin University and the University of Peradeniya have developed the SLIF-Tomato dataset and an AI algorithm to detect tomato diseases in real-world field conditions. The technology aims to provide farmers with a high-accuracy, lightweight tool for early disease identification via mobile devices to reduce crop loss and chemical use.

Propaganda risk 10%

Claims checked 7

Techniques found 1

Topics 3

Coverage spectrum

Coverage gap: Low Left coverage

Left14%

Center72%

Right14%

7 sources compared across this story cluster. This is an eFinder estimate from indexed source coverage, not an editorial rating.

What happened

AI learns to spot tomato diseases using nearly 9,000 field images Sadie Harley Scientific Editor Robert Egan Senior Editor A breakthrough in AI-driven crop disease detection is set to reduce harvest losses and chemical-related health risks, thanks to a…

Why it matters

The Sri Lankan In-Field Tomato (SLIF-Tomato) dataset—built by academics from Charles Darwin University (CDU), the University of Peradeniya (UoP) and others—is the first complete in-field dataset comprising class labels and bounding box annotations collected…

Common ground

The dataset includes 890 raw photos of tomato leaves featuring eight classes: healthy and seven diseases—bacterial spot, early blight, mosaic, powdery mildew, septoria, wilt and late blight—which were then augmented through two phases to produce a total of…

Perspective signals

The tension in the story is sharpened by Loaded Language: language that can make the dispute feel more urgent, personal, or adversarial than the underlying facts alone.

Follow-up questions

What new context would change how readers understand this Agricultural Innovation story?
What evidence would most clearly confirm or weaken the claim that Sri Lanka's tomato yield averaged 18.9 metric tons per hectare as of 2018?
How does this story connect Agricultural Innovation with Sustainable Agriculture over the next few days?

open_in_new Read the original article: https://phys.org/news/2026-08-ai-tomato-diseases-field-images.html

analyticsAnalysis

10%

Propaganda Score

confidence: 95%

Low risk. This article shows minimal use of propaganda techniques.

psychologyPropaganda Techniques Detected

eFinder identified 1 propaganda technique in this article. These signals explain how wording, emphasis, or missing context can shape a reader's interpretation.

warning

Loaded Language 70% confidence

Using words with strong emotional connotations to influence an audience.

Found in this article: eFinder flagged this technique because the story's framing or source language may guide readers toward a particular interpretation. Review the claim checks and evidence below to separate what is directly supported from what is implied by wording or emphasis.

Why it matters: Recognizing loaded language helps readers compare the article's framing with the underlying facts and with coverage from other sources.

fact_checkClaims Checked

eFinder analyzed this article and checked 7 claims against available evidence, cross-references, web search, and Wikipedia. Here is what the fact-checking layer found.

check_circle Corroborated 5

info Single Source 1

verified Verified 1

info

Claim 1: “Sri Lanka's tomato yield averaged 18.9 metric tons per hectare as of 2018”

SINGLE SOURCE

The specific figure of 18.9 metric tons per hectare for Sri Lanka in 2018 is mentioned in one web search result. Other results discuss different countries or general Sri Lankan info without confirming this specific statistic.

menu_book

wikipedia NEUTRAL — Sri Lanka, officially the Democratic Socialist Republic of Sri Lanka, formerly known as Ceylon, is an island country in South Asia. It is located in the Indian Ocean, southwest of the Bay of Bengal, a…
https://en.wikipedia.org/wiki/Sri_Lanka

menu_book

wikipedia NEUTRAL — Sri Lankan Tamils (Tamil: இலங்கை தமிழர், ilankai tamiḻar or ஈழத் தமிழர், īḻat tamiḻar), also known as Ceylon Tamils or Eelam Tamils, are Tamils native to the South Asian island state of Sri Lanka. Tod…
https://en.wikipedia.org/wiki/Sri_Lankan_Tamils

menu_book

wikipedia NEUTRAL — The Sri Lankan civil war was fought in Sri Lanka from 1983 to 2009. Beginning on 23 July 1983, it was an intermittent insurgency against the government by the Liberation Tigers of Tamil Eelam (LTTE, a…
https://en.wikipedia.org/wiki/Sri_Lankan_civil_war

+ 3 more evidence sources

verified

Claim 2: “Romiyal George et al, SLIF-Tomato: Inverted Residual Convolutional Block Attention Module for In-field Tomato Leaf Disease Recognition, Neural Computing and Applications (2026). DOI: 10.1007/s00521-026-12271-0”

VERIFIED

The full citation including authors, title, journal, year (2026), and DOI is explicitly confirmed by a web search result.

travel_explore

web search NEUTRAL — The SLIF-Tomato is the first tomato leaf disease dataset captured under diverse in-field conditions, which offers (1) disease class labels and (2) bounding box annotations for accurate disease area de…
https://link.springer.com/article/10.1007/s00521-026-12271-0

travel_explore

web search NEUTRAL — Romiyal George et al, SLIF-Tomato: Inverted Residual Convolutional Block Attention Module for In-field Tomato Leaf Disease Recognition, Neural Computing and Applications (2026). DOI: 10.1007/s00521-02…
https://phys.org/news/2026-08-ai-tomato-diseases-field-image…

travel_explore

web search NEUTRAL — The Sri Lankan In-Field Tomato (SLIF-Tomato) dataset contains images of healthy leaves and seven major diseases, providing a foundation for developing faster and more accessible tools for farmers. The…
https://www.isaaa.org/kc/cropbiotechupdate/article/default.a…

check_circle

Claim 3: “The dataset includes 890 raw photos of tomato leaves featuring eight classes: healthy and seven diseases—bacterial spot, early blight, mosaic, powdery mildew, septoria, wilt and late blight—which were then augmented through two phases to produce a total of 8,934 images.”

CORROBORATED

Web results explicitly confirm the dataset contains 8,934 images derived from raw photos covering eight classes (healthy and seven diseases).

travel_explore

web search NEUTRAL — Jun 15, 2023 · This is what Apple says about that: From: How to use account recovery when you can’t reset your Apple ID password - Apple Support "In some cases, you might be able to speed up the accou…
https://discussions.apple.com/thread/254930615

travel_explore

web search NEUTRAL — Jul 2, 2025 · Apple Account Security Alert We have detected unusual activity on your Apple ID associated with a transaction at “Apple Store - CA” for $143.95, made via Apple Pay (pre-authorization).
https://discussions.apple.com/thread/256091266

travel_explore

web search NEUTRAL — Aug 24, 2026 · Other information available here at the Apple website: Identify legitimate emails from the App Store or iTunes Store. Recognize and avoid social engineering schemes including phishing m…
https://discussions.apple.com/docs/DOC-250009890

check_circle

Claim 4: “The Sri Lankan In-Field Tomato (SLIF-Tomato) dataset—built by academics from Charles Darwin University (CDU), the University of Peradeniya (UoP) and others—is the first complete in-field dataset comprising class labels and bounding box annotations collected under diverse real-world conditions.”

CORROBORATED

Multiple independent web sources confirm the SLIF-Tomato dataset was built by CDU and UoP and is the first in-field dataset with class labels and bounding box annotations.

travel_explore

web search NEUTRAL — If Barth be a true prophet, he will be a man not only of his own time and place, but of all times and places. If his correction is true, it will have meaning for our own error and false emphasis.
https://ia801403.us.archive.org/9/items/in.ernet.dli.2015.21…

travel_explore

web search NEUTRAL — La Christologie de Karl Barth par Klaas RUNIA Professeur de théologie pratique a la Theologische Hogeschool van de Gereformeerde Kerken à Kampen (Pays-Bas) Barth a souvent été appelé le "théologien de…
https://hokhma.org/articles-pdf/la-christologie-de-karl-bart…

travel_explore

web search NEUTRAL — Le soir du 4 mai 1962, la première question qui lui fut posée avait trait à son christocentrisme : n'a-t-on pas liquidé tout mystère dès lors que l'on identifie, comme le fait Barth, le contenu centra…
https://www.academia.edu/12384912/La_christologie_de_Karl_Ba…

check_circle

Claim 5: “The dataset allows farmers to detect diseases in their crops early and with more than 99% accuracy.”

CORROBORATED

Sources confirm the dataset allows for early detection and reports accuracy rates of 99.66% and 99.91%, which is 'more than 99%'.

travel_explore

web search NEUTRAL — The proposed approach achieved 99.66% and 99.91% accuracy rates on two curated versions of the SLIF-Tomato dataset. Subsequently, the YOLOv12-large model is employed to detect diseased regions, which …
https://link.springer.com/article/10.1007/s00521-026-12271-0

travel_explore

web search NEUTRAL — The Sri Lankan In-Field Tomato (SLIF-Tomato) dataset contains images of healthy leaves and seven major diseases, providing a foundation for developing faster and more accessible tools for farmers.
https://www.isaaa.org/kc/cropbiotechupdate/article/default.a…

travel_explore

web search NEUTRAL — The SLIF-Tomato dataset built by academics from CDU and UoP contains 8,934 images, broken into eight classes. Pictured are examples of their bounding box annotations. The dataset allows farmers to det…
https://www.cdu.edu.au/news/how-9000-photos-built-foundation…

check_circle

Claim 6: “The paper is published in the journal Neural Computing and Applications.”

CORROBORATED

Although the specific evidence section for claim 3 was empty, the evidence for claim 6 explicitly cites the paper as being published in 'Neural Computing and Applications'.

check_circle

Claim 7: “The images were assessed using an AI algorithm called the Inverted Residual Convolutional Block Attention Module (IR-CBAM)”

CORROBORATED

Multiple sources confirm the use of the Inverted Residual Convolutional Block Attention Module (IR-CBAM) for analyzing the images.

travel_explore

web search NEUTRAL — Rule 34 - If it exists, there is porn of it. We have pokemon, my little pony, Other hentai, whatever you want. Serving 15,722,624 posts.
https://rule34.xxx/

travel_explore

web search NEUTRAL — The best Rule 34 of Naruto, Elden Ring, Fortnite, Genshin Impact, FNF, Pokemon, animated gifs, and videos! After all, if it exists, there is porn of it!
https://rule34.us/index.php?r=posts/index

travel_explore

web search NEUTRAL — Rule 34 - If it exists, there is porn of it. We aspire to be the biggest video archive of rule34 content.
https://rule34video.com/latest-updates/

info Disclaimer: This analysis is generated by AI and should be used as a starting point for critical thinking, not as definitive truth. Claims are verified against publicly available sources. Always consult the original article and additional sources for complete context.