使用深度學習，我們很容易地可以分辨圖片的正確性，由於房間室內場景的圖片每一種都很相似，所以需要大量的資料和時間來進行辨識，來提高準確率。因此我們使用遷移學習來改善我們的神經網路，從已經預先訓練好的模組來提取資料，把它遷移到我們的資料中，可以提高我們的資料處理效果，不用花很多的時間來重新訓練一個模組。我們使用Tensorflow 重新訓練 Google的Inception v3 模型，主要分類家、休閒場所、公共空間、商店、工作場所，資料集是來自MIT67和ImageNet，每一種的資料圖片準備六千張以上進行訓練。最終使用圖片進行測試都有50%以上的準確率，使用少量資料，較短的訓練時間。

With deep learning techniques, we can easily distinguish the object the images. However, because the indoor scene images are very similar to each other, so we need a great amount of data and time, to improve classification accuracy. Therefore, we propose using the transfer learning technique to improve our neural network learning. From already trained neural network to extract information in advance, then transfer to other training data. The transfer learning can improve the effect of our data processing, without the need to spend a lot of time to train a module. We use the Tensorflow to retrain the Google's Inception v3 model. The main classification scenes are: Home, Leisure, Public Spaces, Store, and Working Place. The dataset is obtained from the MIT67 and the Imagenet images set. We use 6000 images for each scene above for training. Our experiments show than 50% testing accuracy with a small amount of trainging data, and short training time.

[1] What is the difference between artificial intelligence, machine learning and deep learning?, “https://blogs.nvidia.com.tw/2016/07/whats-difference-artificial-intelligence-machine-learning-deep-learning-ai/”, Referenced on May 22 th, 2018.
[2] Convolutional neural network, CNN, “https://medium.com/@chih.sheng.huang821/%E5%8D%B7%E7%A9%8D%E7%A5%9E%E7%B6%93%E7%B6%B2%E8%B7%AF-convolutional-neural-network-cnn-cnn%E9%81%8B%E7%AE%97%E6%B5%81%E7%A8%8B-ecaec240a631”, Referenced on May 22 th, 2018.
[3] What is Transfer Learning? What are the historical prospects in this field?,
“https://www.pixpo.net/others/K_0068Q0SZ.html”, Referenced on May 22 th, 2018.
[4] Distributed Tensorflow Example - Imanol Schlag, “http://ischlag.github.io/2016/06/12/async-distributed-tensorflow/”, Referenced on May 24 th, 2018.
[5] GoogLeNet Inception(V1-V4), “https://blog.csdn.net/hejin_some/article/details/78636586”, Referenced on May 24 th, 2018.
[6] Pedestrian Detection using Tensorflow and Inception, “http://silverpond.com.au/2016/10/24/pedestrian-detection-using-tensorflow-and-inception.html”, Referenced on May 26 th, 2018.
[7] Hinton. G, Vinyals. O, Dean J, “Distilling the knowledge in a neural network”. arXiv preprint arXiv:1503.02531, 2015.
[8] Rastegari. M, Ordonez. V, Redmon. J, et al. “XNOR-Net: ImageNet Classification Using Binary Convolutional Neural Networks”. arXiv preprint arXiv:1603.05279, 2016.
[9] Wu. J, Leng. C, Wang. Y, et al. “Quantized Convolutional Neural Networks for Mobile Devices”. arXiv preprint arXiv:1512.06473, 2015.
[10] Szegedy. C, Vanhoucke. V, et al. “Rethinking the Inception Architecture for Computer Vision”. In Computer Vision and Pattern Recognition (CVPR), 2016.
[11] Seo. Y, Kyung-shil. S, “Image classification of fine-grained fashion image based on style using pre-trained convolutional neural network”. In Big Data Analysis (ICBDA), 2018.
[12] Krizhevsky. A, Sutskever. I, Hinton. GE, “ Imagenet classification with deep convolutional neural networks ”. In Advances in neural information processing systems, 2012.
[13] Long D. N, Dongyun. L, “Deep CNNs for microscopic image classification by exploiting transfer learning and feature concatenation”. In Circuits and Systems (ISCAS), 2018.
[14] He. K, Zhang. X, Ren. S, Sun J, “Deep Residual Learning for Image Recognition”, In IEEE Conference on Computer Vision and Pattern Recognition (CVPR),2016.
[15] Scott G. J, England M. R, Starms W. A, Marcum R. A, Davis C. H, “Training Deep Convolutional Neural Networks for Land-Cover Classification of High-Resolution Imagery”, In IEEE Geoscience and Remote Sensing Letters, 2017.
[16] Xiaogang. L, Tiantian. P, “Convolutional neural networks based transfer learning for diabetic retinopathy fundus image classification”, In Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI), 2017.
[17] Yian. S, Kyung-shik. S, “Image classification of fine-grained fashion image based on style using pre-trained convolutional neural network”, In Big Data Analysis (ICBDA), 2018.
[18] Rong. T, Lei. W, “Transfer learning for children's speech recognition”, Asian Language Processing (IALP), 2017.
[19] Julius. K, Louis. K, llia. K, et al, “Transfer Learning for Speech Recognition on a Budget”. arXiv preprint arXiv:1706.00290, 2017.
[20] Allah Bux. S, Xiaofeng. W, “Human action recognition using transfer learning with deep representations”, Neural Networks (IJCNN), 2017.