Market Context — Why This Technology, Why Now

Global food demand is rising while arable land and water resources are under increasing strain, exacerbated by unpredictable weather patterns. This drives urgent investment in precision agriculture and smart farming solutions that optimize resource use and enhance resilience. Regulatory pressures for reduced pesticide use and sustainable practices further accelerate the need for technologies like this, which can significantly cut chemical inputs by up to 30% while boosting output.

Key Competitive Advantages
01

Achieves high-precision separation of leaves from background using deep learning, a challenge for conventional methods. Enables early detection of subtle crop stress and optimal environmental control.

02

Reduces physiological disorder and disease risk through combined analysis of leaf temperature, area ratio, duration, and VPD. Contributes to stable yield and quality, potentially reducing waste by up to 20%.

03

Secures strong IP in competitive market, overcoming 10 prior art citations, ensuring market differentiation.

Market Opportunity
Controlled Environment Agriculture & Plant Factories
$150M (AI est.)
High demand for stable production and increased profitability is accelerating the adoption of environmental control technologies. AI-driven precision is the next growth driver.
Large-scale greenhouse operators Vertical farm developers Agricultural technology integrators
Large-Scale Open-Field Cultivation
$200M (AI est.)
Addressing climate change risks and labor shortages are key challenges. This technology could integrate with drones for wide-area crop monitoring and management.
Agribusiness corporations Agricultural drone manufacturers Farm management software providers
Plant Breeding & Agricultural Research
$50M (AI est.)
Plays an essential role as a tool for detailed analysis of crop physiological states in new variety development and cultivation method optimization.
University research departments Seed and agrochemical companies Government agricultural research institutes
IP Defensibility — Why Competitors Can't Replicate This
What This Patent Covers

This patent protects a system and method for precise leaf temperature acquisition and environmental control, specifically covering deep learning-based separation of leaves from background in thermal images and environmental control logic based on leaf temperature and VPD. The patent successfully navigated rigorous examination, demonstrating robust claims and clear differentiation from prior art.

Competitive White Space

This patent primarily covers leaf temperature and VPD-based environmental control. White space exists in integrating advanced soil moisture and nutrient delivery systems, or developing specific pest and disease identification algorithms using alternative spectral imaging beyond thermal.

Economic Impact
~$260K/year estimated yield and quality improvement per facility (est.)
estimated ROI · USD · AI analysis
ROI Calculation Logic

Implementing this technology could reduce waste from physiological disorders and diseases by 8 percentage points (from 10% to 2%) and improve yield quality by 5%. For a controlled environment agriculture farm with $2.0M (AI est.) in annual revenue, waste reduction could yield $2.0M (AI est.) × 8% = $160K (AI est.). Quality improvement could yield $2.0M (AI est.) × 5% = $100K (AI est.). This totals an estimated annual economic benefit of ~$260K (AI est.) per facility, contributing to long-term profitability.

Speed to Market
6× faster than in-house development
This technology benefits from completed fundamental research and algorithm development by a national R&D institution, with the deep learning model foundation already established. This significantly shortens development time compared to building a similar system from scratch. Specifically, the patented leaf-background separation algorithm and VPD-based environmental control logic eliminate the need for licensees to develop these core components, accelerating market entry by an estimated 2.5 years.
Competitive Positioning

X: Precision Control Efficiency
Y: Multifunctionality & Versatility

Business Models & Applications
☁️ SaaS Precision Agriculture Platform
Offers this technology as a cloud-based platform. Users can access leaf temperature data analysis, environmental control algorithms, and growth reports via subscription. This model reduces initial investment, making it accessible to a wider range of farmers.
⚙️ AI-Integrated Environmental Control Hardware Sales
Develop and sell environmental control devices integrating infrared cameras and AI processing units. Easily integrated into existing controlled environment agriculture facilities, enabling high-precision leaf temperature monitoring and automated control on-site.
📊 Agricultural Data Analysis & Optimization Service
Provides detailed analysis of crop growth based on leaf temperature and VPD data obtained by this technology. Offers consulting services to propose optimal cultivation protocols, aiming to maximize yield and improve quality.
Adjacent Application Opportunities
🏥 Medical & Healthcare
Non-Contact Vital Monitoring
Applying leaf temperature acquisition to high-precision human skin surface temperature measurement for non-contact vital monitoring. Deep learning background separation removes clothing and environmental noise, accurately capturing temperature changes. This could support remote medicine, elderly care, and athlete condition management, improving monitoring efficiency by up to 30%.
🏭 Manufacturing & Quality Control
Precision Product Surface Temperature Inspection
Applicable to manufacturing line product surface temperature inspection, automating and enhancing quality control. For example, detecting overheating in electronic components or verifying food cooling states. Deep learning object-background separation enables fast, accurate temperature anomaly detection, potentially reducing defect rates by 25% and improving production efficiency.
⚡ Energy Management
Early Detection of Equipment Anomalies
Could detect abnormal heat generation in large-scale facilities like factories or power plants. Deep learning extracts specific equipment from complex backgrounds, continuously monitoring temperature changes to predict failures. This could prevent major accidents and downtime, potentially reducing maintenance costs by 20% and increasing operational uptime.
Integration Roadmap — Estimated 18-Month Deployment
Phase 1: Technology Validation & Requirements Definition
Duration: 3 months
Analyze the licensee's cultivation environment and crop characteristics to define the technology's scope and specific requirements. Validate integration with existing equipment and data acquisition methods.
Phase 2: System Development & Prototype Construction
Duration: 6 months
Customize the deep learning model and implement environmental control algorithms based on defined requirements. Build a prototype and conduct initial tests in a small-scale demonstration environment.
Phase 3: Pilot Deployment & Full Operation Transition
Duration: 9 months
Initiate full system deployment and data collection in a real-world environment. Fine-tune algorithms based on acquired data, conduct efficacy validation, and optimize performance. Transition to full operation after stable performance is achieved.
Technical Feasibility
This technology can be implemented by integrating deep learning software with off-the-shelf infrared cameras and existing environmental control systems. The patent claims clearly define a thermal image data acquisition unit, a background removal unit (deep learning), and a leaf temperature acquisition unit, structured as software modules easily integrated into existing IoT platforms or control controllers. This offers a technical advantage by enabling high-precision agriculture with minimal capital expenditure, maximizing the use of existing infrastructure.
Success Scenario
Implementing this technology could visualize crop physiological stress in real-time, enabling automated optimal irrigation and ventilation. This has the potential to improve overall yield stability by 20% across the cultivation cycle. Furthermore, by detecting early signs of disease and preventing outbreaks, pesticide usage could be reduced by up to 30%. This would contribute to both stable supply of high-quality crops and reduced environmental impact.
Patent Record
APPLICATION NO.
特願2021-114390
REGISTRATION NO.
7687669
FILING DATE
2021/07/09
GRANT DATE
2025/05/26
EXPIRATION DATE
2041/07/09
PATENT HOLDER
国立研究開発法人農業・食品産業技術総合研究機構
Examination History
2021年09月27日
手続補正書（自発・内容）
2021年11月02日
手続補正書（自発・内容）
2024年04月23日
出願審査請求書
2025年01月23日
拒絶理由通知書
2025年03月07日
手続補正書（自発・内容）
2025年03月07日
意見書
2025年05月02日
特許査定