白帶魚是臺灣近海重要的撈捕和消費的主要魚類之一。臺灣常見的白帶魚品種有三種，目前在撈捕和消費現場要進行白帶魚的品種辨識，都需藉由專家經驗人工辨識，因此難以達到普及應用目的。本研究提出一個白帶魚特徵擷取方法，藉由眼睛中心至頭部輪廓的距離向量，來鑑識白帶魚品種。我們首先透過U-Net神經網路進行白帶魚偵測以及頭部分割，接著提取輪廓以及偵測魚眼，最後計算每隻白帶魚的形狀特徵。為了驗證此一特徵的魚種鑑別性，我們以機率神經網路、決策樹、支援向量機以及k近鄰分類器等四種分類器來進行魚種辨識。

Trichiurus lepturus is one of the main fishes that are harvested and consumed off the coast of Taiwan. There are three common species of Trichiurus lepturus in Taiwan. Currently, fish’s classification at the fishing and consumption sites, manually identified the species of Trichiurus lepturus by experts or experiencers is required. Therefore, it is difficult to achieve the popularization purpose. This research mentions about a feature extraction method for Trichiurus lepturus, which uses the distance vector from the center of the eye to the outline of the head, to identify the species of Trichiurus lepturus. First, we use the U-Net neural network to detect the Trichiurus lepturus and execute the head segmentation, then extract the contours and detect fish eyes, and calculate the shape features of each Trichiurus lepturus. In order to verify the "discriminative" of shape feature, we use four types of classifiers such as probability neural network, decision tree, support vector machine and k-nearest neighbor classifier to classify the fishes.

[1] Mingyuan Fan, Rui Huang, Wei Feng, Jlizhou Sun, “Image blur classification and blur usefulness assessment,” IEEE International Conference on Multimedia & Expo Workshops (ICMEW), pp. 531-536, 2017
[2] Na Zhao, Shi-Kai Sui, Ping Kuang, “Research on image segmentation method based on weighted threshold algorithm,” 12th International Computer Conference on Wavelet Active Media Technology and Information Processing (ICCWAMTIP), pp. 307-310, 2015
[3] Amit Chhabra, “A hybrid approach for color based image edge detection,” International Conference on Advances in Computing, Communications and Informatics (ICACCI), pp. 2443-2448, 2014
[4] R. S. Deepthi, S. Sankaraiah, “Implementation of mobile platform using Qt and OpenCV for image processing applications,” IEEE Conference on Open Systems, pp. 284-289, 2011
[5] Zhao Xu, Xu Baojie, Wu Guoxin, “Canny edge detection based on Open CV,” 13th IEEE International Conference on Electronic Measurement & Instruments (ICEMI), pp. 53-56, 2017
[6] Aylin Sevik, Pakize Erdogmus, Erdi Yalein, “Font and Turkish Letter Recognition in Images with Deep Learning,” International Congress on Big Data, Deep Learning and Fighting Cyber Terrorism (IBIGDELFT), pp. 61-64, 2018
[7] Wei Zhang, Ping Tang, Lijun Zhao, Qingqing Huang, “A Comparative Study of U-Nets with Various Convolution Components for Building Extraction,” Joint Urban Remote Sensing Event (JURSE), pp. 1-4, 2019
[8] Wittaya Koodtalang, Thaksin Sangsuwan, Surat Sukanna, “Glass Bottle Bottom Inspection Based on Image Processing and Deep Learning,” Research, Invention, and Innovation Congress (RI2C), pp. 1-5, 2019
[9] Janak Trivedi, Mandalapu Sarada Devi, Dave Dhara, “OpenCV and Matlab based car parking system module for smart city using circle hough transform,” International Conference on Energy, Communication, Data Analytics and Soft Computing (ICECDS), pp. 2461-2464, 2017
[10] Anand Upadhyay, Santosh Kumar Singh, “Classification of IRS LISS-III images using PNN,” 2nd International Conference on Computing for Sustainable Global Development (INDIACom), pp. 416-420, 2015
[11] Safavian, S. R., & Landgrebe, D., “A survey of decision tree classifier methodology,” IEEE transactions on systems, man, and cybernetics, vol. 21, no. 3, pp. 660-674, 1991
[12] Suykens, J. A., & Vandewalle, J., “Least squares support vector machine classifiers,” Neural processing letters, vol. 9, no. 3, 293-300, 1999
[13] Weinberger, K. Q., Blitzer, J., & Saul, L. K., “Distance metric learning for large margin nearest neighbor classification,” Advances in neural information processing systems, pp. 1473-1480, 2006
[14] O. Ronneberger, P. Fischer, T. Brox, “U-Net: Convolutional Networks for Biomedical Image Segmentation,” International Conference on Medical image computing and computer-assisted intervention, pp. 234-241, 2015
[15] Matheron, G., & Serra, J., “The birth of mathematical morphology,” Proc. 6th Intl. Symp. Mathematical Morphology, pp. 1-16, 2002
[16] Jackway, P. T., & Deriche, M., “Scale-space properties of the multiscale morphological dilation-erosion,” IEEE transactions on pattern analysis and machine intelligence, vol. 18, no. 1, pp. 38-51, 1996
[17] John Canny, “Finding edges and lines in images,” Artif. Intell. Lab., Mass. Inst. Technol., Cambridge, MA, Tech. Rep. Al-TR-720, 1983
[18] 周達峰, 張志永, and 黃遠東, “眼睛檢測演算法的比較,” 國立交通大學電機學院與資訊學院專班，電子與光電學程碩士論文, 2005
[19] Donald F. Specht, “Probabilistic neural networks for classification, mapping, or associative memory,” IEEE International Conference on Neural Networks(ICNN), pp. 525-532, 1988
[20] Donald F. Specht, P. Susie Maloney, “The use of probabilistic neural networks to improve solution times for hull-to-emitter correlation problems,” International Joint Conference on Neural Networks, pp. 289-294, 1989
[21] Shen, Xiaolin, Lu Yu, and Jie Chen, “Fast coding unit size selection for HEVC based on Bayesian decision rule,” picture coding symposium, pp. 453-456, 2012
[22] Emanuel Parzen, “On estimation of a probability density function and mode,” The Annals of Mathematical Statistics, vol. 33, no. 3, pp. 1065-1076, 1962