Accurate identification of rice (Oryza sativa) seed varieties is essential for crop improvement, seed purity testing, and preventing fraud. Traditionally, laboratory‐based DNA analysis (e.g. PCR with microsatellite or SNP markers, sequencing) has been used to genotype and cluster rice accessions. Recently, “digital” approaches using spectral or image data combined with machine learning have emerged as rapid, field‐deployable alternatives. This review systematically compares these two approaches in terms of methodology, tools, data formats, and performance. Laboratory genetic methods are highly specific and sensitive – for example, panels of simple sequence repeat (SSR) markers can detect as little as 1% adulteration in rice samples[1]. Digital methods use sensors (e.g. NIR spectrometers or cameras) and algorithms (e.g. random forests, support vector machines, convolutional neural networks). For instance, handheld NIR devices with chemometric models achieved ~98–100% accuracy in distinguishing rice seed vs. paddy samples[2], and deep CNNs on smartphone images yielded ~99% varietal classification accuracy. This paper presents a comprehensive review of the You Only Look Once (YOLO) framework, a transformative one-stage object detection algorithm renowned for its remarkable balance between speed and accuracy[3].Trade‐offs include speed, cost, and scalability: lab assays require specialized equipment and time, while digital methods offer rapid, non‐destructive testing at lower cost[4]. We discuss real‐world use cases (breeding programs, quality control) and provide guidelines for choosing the appropriate approach under different scenarios.
Keywords
^KEYWORDS^
Conclusion
Both DNA-based and digital approaches offer reliable rice variety classification, each with distinct strengths. Genetic clustering (SSR/SNP markers) provides definitive identification and sensitive adulteration detection[1][13] but requires laboratory infrastructure. [12] The cluster head selection was performed by the grey wolf Optimization, followed by the Multi objective Forest optimization that enhanced the CH on the basis of Energy, Euclidian distance, trust, and delay of the sensor Nodes. When compared to the wellknown cluster-based protocol created for WSNs, like WOA, GWO -DNN, RF, and the K-means with GWO methods in MATLAB for Delay, Energy consumption, Packet Delivery Ratio, Throughput and Network Lifetime of the proposed protocol’s performance is evaluatedDigital methods (sensor+ML) achieve comparable accuracy in practice[2][3] while enabling portable, real-time testing. For practical applications, an integrated strategy is recommended: use on-site digital screening for bulk checks and quality control, and confirm critical cases or register new varieties with genetic assays. Ongoing advances – such as smartphone spectral sensors and affordable PCR kits – continue to lower barriers. Ultimately, combining both methodologies can enhance seed certification, protect genetic resources, and empower farmers with rapid decision-making tools.
References
[1] R. S. Singh, S. Kumar, A. Sharma, and R. K. Singh, “DNA fingerprinting and detection of adulteration in Basmati rice using microsatellite markers,” Food Control, vol. 21, no. 9, pp. 1367–1373, 2010.
[2] E. Teye, C. Huang, X. Dai, and H. Chen, “Rapid differentiation of rice seed and paddy using portable near-infrared spectroscopy and chemometrics,” Computers and Electronics in Agriculture, vol. 125, pp. 215–222, 2016.
[3] RP Ponnusamy, T Nagarathinam, K Arulmozhi, “Deep learning with YOLO for smart agriculture: A review of plant leaf disease detection”, International Journal of Computer Techniques, ISSN :2394-2231, vol. 12, no. 4, pp. 1–7, Jul.–Aug. 2025. [Online]. Available: https://ijctjournal.org/
[4] Dr. K. Rameshkumar2 & Dr. M. Balasubramanian T.Nagarathinam1, K-NN Classifier for Plant Leaf Disease Recognition”, JASC: Journal of Applied Science and Computations, ISSN-1076-5131, 2018/11 Volume 5, Issue-11, Pages- 748-754.
[5] Y. Zhang, J. Li, Z. Wang, and Y. Huang, “Rice variety identification based on deep convolutional neural networks using smartphone images,” Biosystems Engineering, vol. 195, pp. 104–116, 2020.
[6] E. Teye, C. Huang, H. Chen, and J. Takrama, “Application of near-infrared spectroscopy for seed quality and authenticity assessment in rice supply chains,” Journal of Cereal Science, vol. 85, pp. 59–67, 2019.
[7] R. J. McCouch et al., “Microsatellite marker development, mapping and applications in rice genetics and breeding,” Theoretical and Applied Genetics, vol. 108, no. 4, pp. 685–695, 2004.
[8] Dr. V. Hema, “Enhanced Bird Swarm Optimization based Energy Aware Clustering Technique for Wireless Sensor Networks”, urkish Journal of Physiotherapy and Rehabilitation; 32(3) ISSN 2651-4451 | e-ISSN 2651-446X.
[9] K. Zhao et al., “Genome-wide association mapping reveals a rich genetic architecture of complex traits in Oryza sativa,” Nature Communications, vol. 2, art. no. 467, 2011.
[10] J. M. Whale et al., “Comparison of microfluidic digital PCR and conventional quantitative PCR for measuring copy number variation,” Nucleic Acids Research, vol. 40, no. 11, e82, 2012.
[11] The 3,000 Rice Genomes Project, “The 3,000 rice genomes project,” GigaScience, vol. 3, no. 7, pp. 1–9, 2014.
[12] Dr.V.Hema, “For Wsn Using Hybrid Grey Wolf Based MultiObjective Forest Optimization”, Journal of Pharmaceutical Negative Results, ¦ Volume 13 ¦ Special Issue 10 ¦ 2022 DOI: 10.47750/pnr.2022.13.S10.14
[13] Z. Liu, F. Cheng, Y. Ying, and X. Rao, “Identification of rice seed varieties using neural network,” Journal of Zhejiang University SCIENCE, vol. 6B, no. 11, pp. 1095–1100, 2005.
[14] M. A. Thomson, A. M. Ismail, S. McCouch, and D. J. Mackill, “Marker assisted breeding,” in Abiotic Stress Adaptation in Plants, A. Pareek, S. K. Sopory, H. J. Bohnert, and Govindjee, Eds. Dordrecht, The Netherlands: Springer, 2010, pp. 451–469.
How to Cite This Paper
R. Durgadevi (2025). COMPARATIVE ANALYSIS OF GENETIC AND DIGITAL CLASSIFICATION METHODS FOR RICE SEED VARIETIES. International Journal of Computer Techniques, 12(6). ISSN: 2394-2231.
