300 Billion Yuan Pet Economy—An Identity Revolution is Calling
In 2025, the scale of China's pet economy exceeded 300 billion yuan. Pets are no longer merely companion animals, but core members of the family. They travel by high-speed rail, seek medical treatment at hospitals, and purchase exclusive insurance—yet when pets enter the public service chain, a fundamental issue emerges: how to prove "it is it"?
Traditional collar tags are prone to loss or forgery, microchip implantation requires surgery and carries rejection risks, and nose print recognition has limited accuracy... Pet identity management urgently needs a more reliable technical solution.
Homsh has been deeply engaged in the iris recognition field for 14 years. Today, it officially releases the White Paper on the Pet Iris Recognition Industry, systematically sorting out policy opportunities, technical routes, and commercial prospects in this emerging field, while disclosing for the first time the 4th-generation OVAI deep learning algorithm system tailored for pet scenarios.
The Policy Window is Open
2025 marks a breakthrough year for pet travel. The pilot project for pet transportation on high-speed rail started with the Beijing-Shanghai line and has gradually expanded to eight trunk lines including Beijing-Guangzhou, Beijing-Harbin, and Shanghai-Kunming, covering 25 stations and 38 trains. Pets and their owners departing and arriving together are evolving from individual pilot projects to national infrastructure.
A special report by Xinhua News Agency clearly stated: Nose print and iris recognition are regarded as important technical directions for pet identity authentication due to their convenience and accuracy. This is the first time that national-level media has confirmed the strategic value of biometric technology in the pet management field.
The policy signal is clear, and the industrial opportunity window has been opened.
Why Iris Recognition?
The iris is a circular tissue in the eyeball located between the cornea and the lens. Its textural features stabilize at approximately one year old after birth and remain unchanged for a lifetime. Even for identical twins, or even the left and right eyes of the same individual, the iris textures are distinct.
In terms of technical performance, the False Acceptance Rate (FAR) of iris recognition can reach 10⁻⁷, far superior to the 10⁻⁴~10⁻⁵ of nose print recognition. More importantly, iris recognition is non-contact and difficult to forge, making it naturally suitable for scenarios requiring high-confidence verification.
The white paper positions iris recognition as a "highly trusted identity factor" in the pet identity management system—not necessarily the sole identity marker, but indispensable in key links such as shipping handover, insurance claim settlement, and lost and found.
Six Core Scenarios Covering the Entire Pet Lifecycle
The white paper conducts an in-depth analysis of six core application scenarios for pet iris recognition:
1. Transportation and Shipping Chain — From shipping order placement to arrival and collection, iris verification runs through the entire process, eliminating risks of misclaiming, fraudulent claiming, and substitution.
2. Urban Dog Registration and Governance — An upgraded non-contact and tamper-proof method that addresses the loopholes of relying solely on documents or manual verification.
3. Pet Hospitals and Chain Medical Services — Establishes medical records for initial consultations and automatically retrieves records for follow-up visits, providing a reliable identity anchor for the long-term accumulation of health data.
4. Pet Insurance and Risk Control — Implements closed-loop verification across three nodes (insurance application, claim reporting, claim settlement), significantly reducing moral hazards.
5. Lost and Found & Shelter System — Builds a closed loop of loss, shelter, and recovery, ensuring every pet can find its way home.
6. Smart Hardware and Multi-Pet Families — Enables smart feeders to dispense food based on pet identification, smart litter boxes to track health status, and smart access control with precise recognition, forming a technical barrier for high-end product lines.
Technological Breakthrough: 4th-Generation OVAI Deep Learning System
Pet iris recognition faces unique challenges: dogs have approximately elliptical pupils, while cats have vertical slit-shaped pupils, making traditional segmentation algorithms based on circular models struggle to adapt.
Homsh launched its 4th-generation algorithm system in 2024—the OVAI (Optimized Vision AI) deep learning system. Its core innovation lies in an end-to-end optimized design, integrating preprocessing, segmentation, and recognition into a unified neural network model, freeing it from the constraints of prior assumptions about pupil geometry.
The OVAI technical architecture includes three core modules:
● A lightweight convolutional network that extracts multi-scale textural features;
● A Transformer-based adaptive segmentation module that handles complex scenarios such as occlusion, reflection, and pupil deformation;
● An end-to-end recognition module that achieves a leap in system performance through joint optimization.
Experimental data shows that the FAR of OVAI in pet iris recognition tasks can reach 10⁻⁶, representing an order of magnitude improvement compared to traditional algorithms.
Chip-Level Capabilities for Edge Deployment
Technological implementation cannot do without hardware support. Homsh successfully taped out the world's first iris recognition ASIC chip, Qianxin Q80, in 2018:
● Encoding Speed: <0.05 seconds
● Matching Speed: 90,000 times/second
● Average Power Consumption: Only 0.1W
Compared with South Korea's CMITECH EF-45 (6W) and the United States' IRISID iCAM TD100 (7W), the Q80 achieves an energy efficiency advantage of more than 60 times. This ultra-low power consumption feature allows iris recognition modules to be embedded in battery-powered portable terminals and smart pet hardware products.
In Closing
The pet iris recognition industry is standing at a historic opportunity window. At the policy level, the continuous expansion of high-speed rail pet transportation and the revision of urban dog management regulations are advancing; at the technical level, accumulations in algorithms and chips have reached the international leading level; at the market level, the 300 billion yuan pet economy and continuously upgrading service consumption demand provide broad commercial space.
Homsh is willing to work with industrial chain partners to jointly promote the standardization and large-scale application of pet iris recognition technology. We believe that enabling every pet to have a unique, unforgeable, and lifelong digital identity is the best way to safeguard the human-pet bond.
