- 學位論文
應用影像辨識技術來探討最適化固態培養灰白松露生長條件
Optimized Tuber borchii Culture Conditions on the Solid State Fermentation via Application of Image Recognition Technology
摘要
固態基質培養是目前台灣商業化食用真菌培養的主流，本研究分三階段依序建構灰白松露菌（Tuber borchii）最適化固態生長培養基配方的生長模型，第一階段應用商業影像處理軟體及編寫影像數值分析程式來驗證目標圖形面積，驗證結果顯示影像分析面積與真實面積相差0.092 %，可靠性高。第二階段以PDA培養基培養灰白松露菌來驗證時間對菌絲擴散的微生物生長模型，研究結果表示灰白松露菌在PDA上的初始平均接菌面積為47.290 ± 9.532 mm2時，第24天平均生長面積為2640.785 ± 113.215 mm2，平均比生長速率為0.174 ± 0.0114 day-1。最後一階段探討灰白松露菌的最佳比生長速率的固態培養基組成成分，以Box-Behnken實驗設計法設計三階三變培養基配方，反應曲面法擬合比生長速率模型、變異數分析、脊型分析探討和模型驗證。研究結果顯示培養基中的Malt extract與Potato starch對灰白松露菌生長具有顯著性影響，而Glucose則沒有顯著性影響。RSM預測最佳的灰白松露菌比生長速率為0.211 day-1，而實際培養灰白松露菌的比生長速率結果為0.230 ± 0.0172 day-1。顯然培養基預測與實驗結果非常接近。因此，透過數學模型擬合能充分說明其可以作為預測灰白松露菌在Glucose、Malt extract和Potato starch三種培養基組合的比生長速率。
並列摘要
The solid state fermentation is the mainstream of commercial edible fungus cultiva-tion in Taiwan. In this study, the growth model of solid-type growth medium formula was established in three stages. The first step was to use the commercial image processing software, and writing an image analysis program to verify the target graphics area. The results showed the image analysis gave the real area where a difference of 0.092 % with high reliability. In the second step, the fungi growth model of mycelial was verified by culturing Tuber borchii in PDA medium. The results showed that the average area of the inoculated area was 47.290 ± 9.532 mm2 on the PDA. The average growth area was 2640.785 ± 113.215 mm2 on the 24th day, and its average specific growth rate was 0.174 ± 0.0114 day-1. The final step studied the compositions of the solid state medium of the optimized specific growth rate for Tuber borchii. The Box-Behnken experimental design method was used to design the Three-level three-factor for the recipes of culture medium. The response surface methodology was used to fit specific growth rate model, analysis of variance, path of steepest ascent, and model validation. The results showed that malt ex-tract and potato starch in the culture medium had significant effects on the growth of Tuber borchii, but glucose had no significant effect. The RSM predicted the specific growth rate of Tuber borchii was 0.211 day-1, and the specific growth rate of Tuber borchii was 0.230 ± 0.0172 day-1 by experiment. Obviously the prediction and experi-mental results matched very well. Therefore, the mathematic model could be used to fully predict the specific growth rate of the three medium combinations of glucose, malt extract and potato starch.