A Literature Review on Image Processing and Classification Techniques for Agriculture Produce and Modeling of Quality Assessment system for Soybean industry Sample
Bibliographic record
Abstract
Soybean is the most admired and favored food in the world.It is readily available in all the countries.Because of richness in protein, it is recommended and consumed in the daily food.Variety of valueadded edible products can be prepared from Soybean including Soya-Milk, Oil, Pulses etc. for human as well as animals.The countries having major share in Soybean production are United States, Brazil, Argentina, China and India [1].In international food market, the quality of Soybean is a major concerning order to assure the production of best tertiary product by food industry.To ease the trading process the countries like US, Canada have recommended certain criteria and standards to buyers for grading Soybean quality.Thus, the measurement of Soybean quality is essential as well as an equally important requirement in today"s scenario to protect the buyers from substandard produce [2].Accordingly, if the Soybean quality assessment is done manually then the chance of error in measurement is significant and in case it is done with the help of a perfectly trained automatic machine then the chance of error would be minimum.On these aspects, this research work is directed towards the modeling of an automated system for the quality measurements of Soybean Sample [3].In this research work, authors propose a novel two stage model to addresses the issues related to quality assessment of Soybean.A two-stage model comprised of; first stage to detect the physical Abstract: Soybean, the most popular golden bean of America, is widely known for its fat free food products.Richness in Protein makes it one of the best suggested meals which can be consumed in the form of pulses, oil, food for animals etc.The quality of such products is mainly dependent on the quality of Soybean procured as fresh farm produce.Governments and regional authorities have already defined the standards for quality assessment and grading of Soybean, which are meant to be followed in the commercial market while trading.Presently, visual inspection is the preferred way to conduct physical quality assessment of Soybean and it is performed by an expert person, at the time of procurement of Soybean based on the standards recommended by the buying authority.Physical parameters of Soybean kernel like color, growth corresponding to size, damage, fungi/ disease as well as mixing of other material/ objects in soybean sample influence the grade of that sample during quality assessment.Dependency of this process on human expert usually harms the accuracy in assessment.Therefore, an automated machine is desired to be a suitable solution to address this issue and can benefit in terms of increased accuracy, reliability, and reduced response time.Worldwide researchers are working on designing such automated systems for different type of fruits, grains etc.However, in case of Soybean, up till now we are successful in cleaning, sorting, color detection, and also, through various image processing techniques and algorithms researchers could detect the anomalies present in grain sample.But an automated system for the quality assessment and grading of Soybean according to an International Standard is yet to be implemented. In this paper, we propose a two-stage model for the quality assessment of Soybean; first stage focuses on image processing techniques like Image acquisition, preprocessing and feature extraction Likewise the second stage works on classification of Soybean kernels and sample grading with the help of a machine learning technique on the basis of an International Standard.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.003 | 0.004 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.005 | 0.003 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".