SnT NIRWatchdog is a product with the ability to solve real problems of clients in various industries, specifically for authentication of products’ raw materials susceptible to economically motivated tampering. Based on the competitive analysis performed, it was found that SnT NIRWatchdog is ahead of other NIR software solutions. Four main competitive advantages of the SnT NIRWatchdog were identified, which should be leveraged by the team in the next steps: Use of Blockchain Technology, Working Prototype, Device-Agnostic, and Automated Storage.
The NIRWatchdog software platform combines several modules to provide the ability to identify the deviation of a material from the standard material in a given population. The system uses as an input reading from a Near-Infrared (NIR) Spectrometer, and through machine learning provides a prediction result determining if a product deviates from normal. For example, a common use case includes using NIRWatchdog on commercial packaging materials. If the quality of the material has degraded or is subject to fraud, the system would identify it as deviating from the standard range acceptable for its class, and flag it for further scrutiny.
The process begins with a scan using a Near-Infrared Spectrometer, a handheld tool that takes a rapid, highly informative, scan of any item. Our software platform (NIR Watchdog) gets an input reading from a Near-Infrared (NIR) Spectrometer and using the machine learning AI-algorithms it builds the molecular fingerprint of the scanned product.
Machine Learning Technology
The technology can be described in 2 stages: training and inspection.
Training (Stage 1): During this stage, a Machine Learning (ML) model is autonomously trained using data from a genuine product. By applying a series of filters (ML pipeline) of each data sample, a unique signature of the sample is obtained.
Inspection (Stage 2): The flow starts with a user who, having a scanner and the mobile app, scans the requested product. The scan results are sent to the comparison algorithm which determines with a certain probability whether the scanned product could belong to the trained samples’ family. For example, the system can distinguish the fake from genuine products. Then, blockchain stores the results of inspection and location along with the hash value of the data. The data itself is stored in the encrypted database.
User Mobile App
We have developed a mobile app to simplify the user experience when scanning products. The app can be used in both stages previously described. Once the data is scanned, the sample is sent to the server and stored in the database. In the first stage, the data is used to train the system. In the second stage, the user will be notified by the app about the product signature (genuine or not). The response is identified by the timestamp and the ID of the sample, and the input file is defined as CSV and sent to an email.
Using Blockchain Technology and its Potential Influence for Regulatory Adoption
Blockchain is a public database that records data as chained blocks. It is essentially a decentralized architecture built on a multi-party consensus based on multiple peers. Peers in the blockchain are primarily used to endorse legitimate transactions and record them on blockchain ledgers. Blockchain records data through transactions and ensures data consistency through cryptography.
The data acquired by conventional spectrometers can be tampered with; thus, the reliability of data from other spectrometers is low and the data credit is very low. This problem is solved by the NIR Watchdog system by using blockchain technology. Specifically, each spectral data is bound to the spectrometer used by a digital signature. Blockchain records tag information related to the spectral data, such as hash value, signature, timestamp, and system ID. The authenticity of spectral data is essential if the testing is applied to regulatory uses, such as food inspection.
NIR Watchdog Application in Practice
Our research indicates that there is high potential applicability of the developed technology for both commercial and governmental use. The commercial use of the system has a wide range of applications for various industries and services, such as:
Low quality (imitation, counterfeit) detection
Food quality inspection for regulatory purposes in Food industries, such as chocolate, caviar, alcohol/spirits (especial mid-level Vodkas – Finlandia, Smirnoff, Absolut), and others
Luxury cosmetics, clothes, and accessories
Art – Paintings, sculptures, rare coins
Agriculture, in determining optimal harvest timing by assessing fruit maturity
Animals’ nutrient analysis such as Instant check for the quality of animals’ food
Currently, our proof of concept has achieved accuracy results of 98% in detecting counterfeit products.
Competitive Landscape Analysis
Few companies provide near-infrared spectroscopy solutions as in-depth as NIRWatchdog. In many cases, companies primarily sell scanners and software, but do not provide personalized data analysis, nor a Blockchain backend.
SnT NIRWatchdog is currently identified as a leading solution.