Pests are caught in traps, AI identifies their photos into more than 148 species, and the farmer records them in a smartphone pest-control logbook. "VietFruitRise," jointly advanced by Vietnam and Australia, has begun running this whole flow with more than 100 smart traps. How far is field pest control automated, and where does the human hand remain? For Japanese food buyers and growing regions, it becomes an implementation example that foretells how machine-readable the traceability of fruit and rice exported to Japan will become.
The starting point is an on-the-ground report carried by the Vietnamese agricultural media "Nong nghiep Moi truong" on August 8, 2026. Trials are underway at 19 cooperatives across six provinces, including Dong Thap and Lam Dong. Confirming the figures one by one, this article lays out the mechanism and the lessons for Japan.
Lured with pheromones and LED, AI tells apart 148 species
At the center is RYNAN's smart pest trap called "InSentinel." It lures insects with LED light and pheromones and automatically photographs the individuals that enter the trap. The mechanism is that AI analyzes those images, determines the species and count on the spot, and notifies an app called "MEKONG."
The insects it can identify number more than 148 species. It distinguishes not only rice and fruit-tree pests such as planthoppers and leaf-rolling insects but even the predatory natural enemies that eat them. Whether pests and beneficial insects can be counted without being confused bears directly on the decision of whether to spray. The design is for species identification to stop the waste of preventively spraying a field that already has enough natural enemies.
The University of Queensland in Australia joined the development, with Dr. Vo Doan Thang and emeritus professor Robert Henry named among the participants. Dr. Thang places the local aim on the point that pests and diseases have become harder to read amid erratic weather. With the real data the traps gather, they are trying to replace outbreak forecasting that had relied on intuition. Henry has said that the two countries' smart-agriculture cooperation will drive the needed technological innovation, and the picture of combining Vietnam's field data with Australia's research base comes into view. The design of not ending the trap as a standalone measuring instrument but connecting its determination accuracy to international research becomes a foothold for a late-arriving growing region to raise its level all at once.
The pest-control logbook cannot be fully automated: the human hand that remains there
What we do not want to overlook here is that the logbook itself is not fully automated. Records of work such as spraying and fertilizing are entered by the farmer typing text into the app or reading it aloud by voice. Even if the trap's determination is automatic, a person records what was done in the field.
AI automatically cross-checks the entered data against standards (VietGAP, GlobalGAP, and the importer's requirements) and warns if a violation risk emerges, such as insufficient pre-harvest interval days. What is automatic is the "cross-checking and warning," not the "recording." Confusing this places unreasonable expectations on the ground. Providing voice input is an adaptation to the reality of fieldwork, where hands get dirty and the time to write is grudged.
VietFruitRise's implementation scale in figures
| Item | Figure |
|---|---|
| Smart traps installed | Over 100 units |
| Insect species AI identifies | Over 148 species |
| Cooperatives taking part in the trial | 19 cooperatives |
| Preparation and processing time for certification documents | Reduced by about 40% |
| Provinces of deployment | Six provinces (Dong Thap, Long An, Vinh Long, Lam Dong, Phu Yen, Nghe An) |
Source: Nong nghiep Moi truong (August 8, 2026). Figures are approximate, based on the original article's statements.
The place of "V-STANDARD," which cuts certification paperwork by 40%
Beyond the traps and the app is a certification-support mechanism called "V-STANDARD." It digitizes the preparation of certification documents that used to be bundled on paper and compiles them automatically from the farmer's entered data. The figure of cutting preparation and processing time by about 40% refers to this reduction around the paperwork. The electronic format takes over the verification work that auditors used to spend weeks cross-checking.
The exit VietFruitRise aims for is to get rice and fruit through to demanding markets like the EU and Japan. Count pests with traps, record work in the app, and compile the documents with V-STANDARD. Only when the three connect does a growing region move from "having records" to "being able to prove." Vietnam's growing-region DX aiming for a fourfold unit price, the smart-enabled dragon fruit of northern Phu Tho province, faces the same direction as such efforts while placing its weight on pest control and certification.
What differs from Japan's outbreak forecasting and precision pest control
Japan too has tools for catching pests rooted in practice. Pheromone traps are widely used in paddy rice and fruit trees, and the prefectural pest-control stations have handled outbreak forecasting and built spraying calendars. It is not that Vietnam's implementation leads in individual technologies.
Where the difference shows is in how things are connected. In Japan, trap data, farmers' work records, and certification and shipping documents tend to be scattered across separate systems. Traps are counted by eye on-site, logbooks are on paper or in disparate apps, and certification documents are prepared separately again. Outbreak forecasting is aggregated at the prefectural level and is highly accurate, but the idea of carrying per-field determinations through to the farmer's pest-control records and shipping documents in one continuous line is thin. VietFruitRise is trying to put everything from the trap's automatic determination through the logbook to the certification documents onto a single line. Reducing pests to raise the unit price, the reduced-pesticide rice of Quang Ninh province and, combining drones with alternate wetting and drying, the rice-farming DX of An Giang province—as with these, Vietnam's strength lies not in individual technologies but in a design that "runs them together."
Precisely because it is a late arrival, it can build the system from the start on the premise of export proof. This difference in order tells on how machine-readable the documents for export to Japan arrive.
Points to watch from now on in sourcing for Japan
- Whether the trap data, the pest-control logbook, and the certification documents are connected in the same system. If they are fragmentary, gaps in proof remain.
- As long as logbook entry is left to the farmer, how are entry gaps filled? Confirm the actual operation of voice input and whether the warning for unentered items functions.
- Whether the identification of 148 species covers the main pests of the target fruit and rice. Pests differ by crop, and a change in tree species can cause oversights.
- Whether the figure of about 40% reduction applies to the formats and inspection items your company requires. A reduction premised on VietGAP or GlobalGAP does not necessarily cover the Japanese side's individual requirements.
The scale of 19 cooperatives at the trial stage is still too small to represent the whole growing region. As a buyer, you need the perspective of evaluating separately the cooperatives that have implemented it and those that have not.
What remains once the paper pest-control logbook disappears
Smart traps tell apart 148 species, more than 100 units operate across six provinces, and they cut certification-document time by about 40%. Up to here are verifiable facts. On the other hand, the logbook recording itself still rests in human hands, and the trial scale of 19 cooperatives is not the full picture of the growing region. It is at a stage that should be read without seeing it as either too much or too little.
What Japanese growing regions can learn is the order of connecting traps, logbooks, and certification into a single line, rather than increasing individual smart devices. What those involved in sourcing for Japan should ask is not the novelty of the equipment. Whether that pest data reaches the certification documents by machine—the quality of proof is determined by how the system is connected.
