今後の課題

今後取り組むべき課題として，次の4点が挙げられる．

≤初期視差計算が不要なステレオマッチング回路の実現

 提案した2種類のステレオマッチング回路は，いずれも回路での処理とは別に初期視差を 算出するフェーズを設けており，かつ回路での処理と同等の時間をその処理に要している．更 なる高速化やハードウェア化の容易さを考慮すると，今後は回路の処理だけでステレオマッ チングを実現する必要がある．

≤ハードウェア指向のCENの提案

 ハードウェア化を考慮した自動構築法の検討として，回路規模の縮小と高速化，メモリの 削減，パイプライン処理の3点を実現するための制約条件について述べた．これらの他，今 後はゲート遅延や消費電力等の制約条件を評価関数に取り入れ，よりハードウェア化向きの 回路の自動構築法を提案する．

≤FPGAボード上への実装と検証

 超解像処理回路のハードウェア化として，Verilog-HDLを用いて自動構築した超解像処理 回路の設計とシミュレーションまでを行った．今後は設計した回路をFPGAボード上に実装 して動作させ，シミュレーション結果との比較，検討を行う必要がある．

≤HDL自動生成による，回路自動構築からハードウェア化までの自動化

 現状は最適化された回路構造を人手によってHDLで記述している．今後，回路構造をHDL として生成させることによって回路構築からハードウェア化までを自動化することができる．

謝辞

博士課程後期進学の後押しをして頂き，本研究を進めるにあたり終始多大なるご指導ご助言，豊 かな研究環境を賜りました長尾智晴先生に深く感謝致します．また，本論文をまとめるにあたり貴 重なご指導，ご助言を頂きました田村直良先生，森辰則先生，岡嶋克典先生，富井尚志先生に感謝 致します．

そして，博士号取得を志すきっかけを頂きました國分泰雄先生，貴重なご意見を頂きました長尾 研究室の皆様と先に卒業された先輩や同期の皆様，博士課程後期進学を許可して頂きました人事本 部の皆様，博士号取得を応援して下さったSPT42の皆様に感謝の意を表します．

最後に，家族や友人をはじめ，温かく見守ってくれた方々に感謝します．ありがとうございま した．

参考文献

［1］L.O. Chua and L. Yang. Cellular neural network: Theory. IEEE Trans. Circuits and Systems, Vol. 35, No. 10, pp. 1257–1272, 1988.

［2］L.O. Chua and L. Yang. Cellular neural network: Applications.IEEE Trans. Circuits and Systems, Vol. 35, No. 10, pp. 1272–1290, 1988.

［3］Takeshi Agui, Tomoharu Nagao, and Hiroshi Nagahashi. A genetic method for optimization of asynchronous random neural networks and its application to action control. InProc. Int. Joint Conf. on Neural Networks (IJCNN ’93), Vol. 2, pp. 1893–1896, Nagoya, Japan, 1993.

［4］白川真一,長尾智晴. RFCNによる連続値空間上での自律エージェントの行動制御.電学論C, Vol. 127, No. 5, pp. 762–769, 2007.

［5］Junji Otsuka, Noriko Yata, and Tomoharu Nagao. Cellularly connected evolutionary neural net-works and its application to image processing.Proc. Int. Workshop on Advanced Image Technology (IWAIT ’10), p. 9, 2010.

［6］大塚純二,矢田紀子,長尾智晴. 進化型ニューラルネットワークのセル結合モデルによる画像 変換.電気学会論文誌C, Vol. 3, pp. 430–438, 2012.

［7］大塚純二,矢田紀子,長尾智晴.進化型神経回路網のセル結合モデルによる画像分類.計測自動 制御学会第37回知能システムシンポジウム, pp. 1–6, 2010.

［8］G. Crist´obal, E. Gil, F. ˇSroubek, J. Flusser, C. Miravet, and F. B. Rodr´ıguez. Superresolution imaging: a survey of current techniques. InProc. Advanced Signal Processing Algorithms, Archi-tectures, and Implementations XVIII, Vol. 7074, pp. 7074C–1–7074C–18, 2008.

［9］J.D. van Ouwerkerk. Image super-resolution survey. Image and Vision Computing, Vol. 24, pp.

1039–1052, 2006.

［10］Xin Li and Michael T. Orchard. New edge-directed interpolation. IEEE Trans. Image Processing, Vol. 10, No. 10, pp. 1521–1527, 2001.

［11］S. Battiato, G. Gallo, and F. Stanco. Smart interpolation by anisotropic diffusion. InProc. 12th Int.

Conf. on Image Analysis and Processing, pp. 572–577, 2003.

［12］S. Battiato, G. Gallo, and F. Stanco. A locally-adaptive zooming algorithm for digital images.

Image Vision and Computing Journal, Vol. 20, No. 11, pp. 805–812, 2002.

［13］M. Irani and S. Peleg. Motion analysis for image enhancement: Resolution, occlusion and trans-parency.Journal of Visual Communication and Image Representation, Vol. 4, No. 4, pp. 324–335, 1993.

［14］Raanan Fattal. Upsampling via imposed edges statistics. ACM Trans. Graphics (Proceedings of SIGGRAPH 2007), Vol. 26, No. 3, pp. 95–102, 2007.

［15］Jian Sun, Zongben Xu, and Heung-Yeung Shum. Image super-resolution using gradient profile prior. InProc. IEEE Conf. on Computer Vision and Pattern Recognition (CVPR 2008), pp. 1–8, 2008.

［16］Yu-Wing Tai, Shuaicheng Liu, Michael S. Brown, and Stephen Lin. Super resolution using edge prior and singale image detail synthesis. InProc. IEEE Conf. on Computer Vision and Pattern Recognition (CVPR 2010), pp. 2400–2407, 2010.

［17］W.T. Freeman, T.R. Jones, and E.C. Pasztor. Example-based super-resolution. IEEE Computer Graphics and Applications, Vol. 22, No. 2, pp. 56–65, 2002.

［18］Qi Pan, Chengying Gao, Ning Liu, and Jiwu Zhu. Single frame image super-resolution based on sparse geometric similarity. Journal of Information and Computation Science, Vol. 7, No. 3, pp.

799–805, 2010.

［19］Hong Chang, Dit-Yan Yeung, and Yimin Xiong. Super-resolution through neighbor embedding.

InProc. IEEE Conf. on Computer Vision Pattern Recognition (CVPR2004), Vol. 1, pp. 275–282, 2004.

［20］D. Glasner, S. Bagon, and M. Irani. Super-resolution from a single image. InProc. IEEE Interna-tional Conference on Computer Vision (ICCV2009), pp. 349–356, 2009.

［21］C.-Y. Yang, J.-B. Huang, and M.-H. Yang. Exploiting self-similarities for single frame super-resolution. InProc. Asian Conference on Computer Vision (ACCV 2010), pp. 1807–1818, 2010.

［22］NVIDIA Corporation. Cg toolkit user’s manual. http://developer.nvidia.com/cg-toolkit/.

［23］Labeled faces in the wild database. http://vis-www.cs.umass.edu/lfw/.

［24］Caltech buildings dataset. http://www.vision.caltech.edu/malaa/datasets/caltech-buildings/.

［25］Z. Wang, A.C. Bovik, H.R. Sheikh, and E.P. Simoncelli. Image quality assessment: from error visibility to structural similarity. IEEE Trans. Image Processing, Vol. 13, No. 4, pp. 600–612, 2004.

［26］佐藤浩,小野功,小林重信.遺伝的アルゴリズムにおける世代交代モデルの提案と評価.人工知 能誌, Vol. 12, No. 5, pp. 734–744, 1997.

［27］T. Kanade and M. Okutomi. A stereo matching algorithm with an adaptive window: theory and experiment. Vol. 16, No. 9, pp. 920–932, 1994.

［28］K. J. Yoon and I. S. Kweon. Adaptive support-weight approach for correspondence search. Vol. 28, No. 4, pp. 650–656, 2006.

［29］Minglun Gong and Yee-Hong Yang. Fast stereo matching using reliability-based dynamic program-ming and consistency constraints. InProc. IEEE Conf. on Computer Vision and Pattern Recogni-tion (ICCV ’03), Vol. 1, pp. 610–617, 2003.

［30］H. Hirschm¨ueller. Accurate and efficient stereo processing by semi-global matching and mutual information. InProc. IEEE Conf. on Computer Vision and Pattern Recognition (CVPR ’05), Vol. 2, pp. 807–814, 2005.

［31］X. Sun, X. Mei, S. Jiao, M. Zhou, and H. Wang. Stereo matching with reliable disparity propa-gation. InProc. IEEE Int. Conf. on 3D Digital Imaging, Modeling, Processing, Visualisation and Transmission (3DIMPVT), pp. 132–139, 2005.

［32］Y. Boykov, O. Veksler, and R. Zabih. Fast approximate energy minimization via graph cuts. Vol. 23, No. 11, pp. 1222–1239, 2001.

［33］Yasuyuki Matsushita, Daisuke Miyazaki, and Katsushi Ikeuchi. Interactive shadow removal from a single image using hierarchical graph cut. InProc. Asian Conf. on Computer Vision ’09, Vol. 1, pp. 234–245, 2009.

［34］Y. Boykov and Vladimir Kolmogorov. An experimental comparison of min-cut/max-flow algo-rithms for energy minimization in vision. Vol. 26, No. 9, pp. 1124–1137, 2004.

［35］N. N. Zheng, J. Sun, and H. Y. Shum. Stereo matching using belief propagation. InProc. European Conf. on Computer Vision ’02, Vol. 2, pp. 510–524, 2002.

［36］R. Yang, S. Wang, M. Liao, Q. Yang, L. Wang, and D. Nist´er. Real-time global stereo matching using hierarchical belief propagation. InProc. British Machine Vision Conference ’06, pp. 989–

998, 2006.

［37］P. Felzenszwalb and D. P. Huttenlocher. Efficient belief propagation for early vision. InProc. IEEE Conf. on Computer Vision and Pettern Recognition (CVPR ’04), Vol. 1, pp. 261–268, 2004.

［38］D. Schatstein and R. Szeliski. Middlebury stereo vision page. http://vision.middlebury.edu/stereo/.

［39］A. Klaus, M. Sormann, and K. Karner. Segmented-based stereo matching algorithm using be-lief propagation and a self-adapting dissimilarity measure. Proceedings of the 18th International Conference on Pattern Recognition, Vol. 3, pp. 15–18, 2006.

［40］Z. Wang and Z. Zheng. A region based stereo matching algorithm using cooperative oprimization.

IEEE Conference on Computer Vision and Pattern Recognition, pp. 1–8, 2008.

［41］Q. Yang, L. Wang, R. Yang, H. Ste´enius, and D. Nist´er. Stereo matching with color-weighted correlation, hierarchical belief propagation, and occlusion handling. IEEE Transactions Pattern Analysis and Machine Intelligence, Vol. 31, No. 3, pp. 492–504, 2009.

［42］D. Scharstein and C. Pal. Learning conditional random fields for stereo. InProc. IEEE Conf. on Computer Vision and Pattern Recognition (CVPR ’07), pp. 1–8, 2007.

［43］H. Hirschmuller and D. Scharstein. Evaluation of cost functions for stereo matching. InProc. IEEE Conf. on Computer Vision and Pattern Recognition (CVPR ’07), pp. 1–8, 2007.

［44］J.Canny. A computational approach to edge detection.IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 8, No. 6, pp. 679–698, 1986.

［45］Julian F Miller and Peter Thomson. Cartesian genetic programming. InProc. the European Con-ference on Genetic Programming, pp. 121–132, 2000.

［46］日本アルテラ株式会社. Quartus II web edition. http://www.altera.co.jp/products/software/ quartus-ii/web-edition/qts-we-index.html.

［47］Mentor Graphics Corp. ModelSim. http://www.mentorg.co.jp/products/fv/products/modelsim.html.

［48］日本アルテラ株式会社. TimeQuest timing analyzer. http://www.altera.co.jp/products/software /products/quartus2/verification/time-quest/tq-index.html.