高速魚種識別システムにおける画像分割処理アルゴリズムの効率化について
スポンサーリンク
概要
- 論文の詳細を見る
A new color image processing system with multi-DSP is constructed in practice and applied to fish pattern recognition. In this application, the shapes of fish are very similar but the colors and/or patterns of which are different. The proposed system has a function of variable partition of a picture. Therefore, the size, position and number of processing regions can be varied flexibly and dynamically for each processing task. As a result, it is clarified that the color texture analysis such as histogram and co-occurrence matrix is useful in fish pattern recognition and that the recognition time less than 250msec. per a fish is achieved.
- 長崎大学工学部の論文
- 1994-07-00
長崎大学工学部 | 論文
- パ-ソナルコンピュ-タによる熱物性値プログラム・パッケ-ジの開発--水および空気を含む13種類の理想気体
- THE DIFFUSION AND TRAPPING OF HYDROGEN IN COLD WORKED NICKEL
- ポリエ-テル系ポリウレタン網目の13C-NMRによるキャラクタリゼ-ション--原料・モデルウレタン・ウレア・アロハネ-ト
- セグメンティドポリウレタンの動的粘弾性への鎖延長剤の影響
- 建築構造物の応答制御設計法に関する研究-1-制震構造物の振動実験