查詢結果分析
來源資料
相關文獻
- 電腦類神經網路應用於糙米品質檢測之研究
- 胴裂米以類神經網路檢測之研究
- 糙米品質自動檢測分級系統(2)--糙米外觀品質檢測之研究
- 稻米品質自動檢測分級系統
- 糙米品質自動檢測分級系統--(1)檢測系統之研製
- 應用影像視覺於超薄型表面載式電感器之瑕疵品辨識
- Neural Network Procedures for Taguchi's Dynamic Problems
- A Fast and Efficient Competitive Learning Design Algorithm Based on Weight Vector Training in Transform Domain
- 專家系統振動訊號圖型判別之研究
- 反傳遞模糊類神經網路於流量推估之應用
頁籤選單縮合
題 名 | 電腦類神經網路應用於糙米品質檢測之研究=The Study of Quality Inspection of Brown Rice by Computer Neural Networks |
---|---|
作 者 | 萬一怒; 楊智超; | 書刊名 | 農林學報 |
卷 期 | 46:4 1997.12[民86.12] |
頁 次 | 頁59-81 |
分類號 | 434.1 |
關鍵詞 | 類神經網路; 稻米品質; 機械視覺; 自動檢測; Neural network; Rice quality; Machine vision; Automatic inspection; |
語 文 | 中文(Chinese) |
中文摘要 | 本文探討結合機械視覺與類神經網路對糙米品質檢測,實驗以臺農67號糙米中屑 米、變色粒、發芽粒、褐色粒、異型粒、碎粒、白堊質粒、畸型粒、未卜熟粒及完整粒為樣 本,以影像處理擷取16項參數經正規化後輸入不同類神經網路學習,測試結果顯示完整粒與 人工分類相符率超過 94%、白堊質粒 80% 左右、發芽粒 85% 以上、碎粒 84%、屑米 79.9% 、 未熟粒 75% 以上、異型粒 64%、褐色粒為 67%、變色粒 30% 左右、畸型粒只達 29.67% 較不理想。研究亦顯示了現今以主觀方式設定,非量化的檢測標準中有許多並不盡合理的地 方,而以機械視覺量化檢測有其客觀性。 |
英文摘要 | Machine vision and computer neural networks are applied in this research for brown rice quality inspection. Experiment is based on "Tainung 67" brown rice. The screening, discolored, sprouted, rusty, off-type, broken, chalky, abnormal, immature and sound kernels are selected by man to use as learning and testing samples. First, image processing technique is applied to extract 16 parameters from the kernels. These parameters are normalized and then sent to train several neural networks. The tests show the average categorized ratio of sound kernel by neural networks is over 94%. The ratio of chalky kernel is around 80%,sprouted kernel is around 85%, broken kernel is above 84%, screening is 79.9%,immature kernel is above 75%, off-type kernel is 64%, rusty kernel is 67%, discolored kernel is 30%, and abnormal kernel is only 29.7%. From the study, it shows that the current non-quantify standard of brown rice inspection proposed subjectively from eyes is not proper for the application of machine vision. The quality inspection of brown rice by using machine vision can achieve objective inspection. |
本系統中英文摘要資訊取自各篇刊載內容。