Vol.39 No. 8 August 2017 Hyeonwoo Noh [4] boundng box PASCALV VOC PASCAL VOC Ctyscapes bt 8 bt 1 14 bt

Size: px

Start display at page:

Download "Vol.39 No. 8 August 2017 Hyeonwoo Noh [4] boundng box PASCALV VOC PASCAL VOC Ctyscapes bt 8 bt 1 14 bt"

面莫
5 years ago
Views:

1 39 8 Vol.39 No Infrared Technology August ,2,3 1,3 1, TP A (2017) Infrared Scene Understandng Algorthm Based on Deep Convolutonal Neural Network WANG Chen 1,2,3 TANG Xny 1,3 GAO Sl 1,3 (1. Shangha Insttute of Techncal Physcs, Shangha , Chna; 2. Unversty of Chnese Academy of Scences, Bejng , Chna; 3. Key Laboratory of Infrared System Detecton and Imagng Technology, Chnese Academy of Scences, Shangha , Chna) AbstractWe adopt a deep learnng method to mplement a semantc nfrared mage scene understandng. Frst, we buld an nfrared mage dataset for the semantc segmentaton research, consstng of four foreground object classes and one background class. Second, we buld an end-to-end nfrared semantc segmentaton framework based on a deep convolutonal neural network connected to a condtonal random feld refned model. Then, we tran the model. Fnally, we evaluate and analyze the outputs of the algorthm framework from both the vsble and nfrared datasets. Qualtatvely, t s feasble to adopt a deep learnng method to classfy nfrared mages on a pxel level, and the predcted accuracy s satsfactory. We can obtan the features, classes, and postons of the objects n an nfrared mage to understand the nfrared scene semantcally. Key wordsnfrared magesnfrared scenesemantc segmentatonconvolutonal neural network 0 Jonathan Long CVPR 2015 FCN [1] S. Zheng ICCV2015 CRFasRNN [2] condtonal random feldcrf PASCAL VOC ICLR2015 Lang-Cheh Chen FCN Deeplab [3] CRF E-mallkame@sna.com 2011xcwzk

2 Vol.39 No. 8 August 2017 Hyeonwoo Noh [4] boundng box PASCALV VOC PASCAL VOC Ctyscapes bt 8 bt 1 14 bt [I mn, I max ] 2[0,255] I mn f In Imn In Imn Io 255 f Imn In Imax (1) Imax I mn Imax f In Imax 3 CLAHE [5] 8 bt bt.mat8 bt.jpg.png.png (a) (a) Infrared mage 1 (b) (b) Ground truth Fg.1 Sample of nfrared mage datasetss 729

2 Comparson of compresson results 2 2.1 Softmax 3 2.

3 39 8 Vol.39 No Infrared Technology August 2017 (a) (b) (a) Compresson algorthm based on global hstogram (b) Our algorthm 2 Fg.2 Comparson of compresson results Softmax VGG-16 VGG-16 [6] ImageNet ILSVRC Imagenet AlexNet GoogleNet 2.3 VGG Hole FCN [3] Hole FCN VGG Hole Fg.3 Semantc segmentaton algorthm flow 1 VGG-16 Table 1 VGG-16 framework VGG-16 conv3-64 conv3-64 conv3-128 conv3-128 conv3-256 conv3-256 conv3-256 fc-4096 fc-4096 fc-1000 soft-max 730

4 39 8 Vol.39 No August hole Fg.4 Illustraton of the hole algorthm CRF [7] (2) E( x) ( x ) ( x, x ) j j j x (x )lgp(x ) P(x ) DCNN (, ) (, ) (, ) (3) x x K m j j x x k f f j m 1 m j x x j (x, x j )1 0 k m j m p p I I p p exp( ) exp( ) 1 j j j (4) CRF Ctyscapes [8] Ctyscapes VGG-16 Ctyscapes 54 mn-batch ms/frame Ubuntu14.04 Caffe CUDA GPU NVDIA GM G IUn j jn cl t j n j [1] n / t (5) (1/ n ) n / t cl IU (1/ n ) n /( t n n ) IU cl j j (6) (7) (8) 1 ( tk) tn /( t nj n) k j

5 Infrared Technology Vol.39 No.8 August Table 2 Comparson of predcton accuracy IU IU 1 fc ctyscapes crf ctyscapes fc crf fc crf IU Table 3 IU results of each class on IR dataset IU IU IU 3 fc crf (a) (b) (a) Orgnal mage (b) Labels mage background person 5 (c) (c) Coarse predcton results (d) CRF (d) CRF refne resultss vehcle buldng tree Fg.5 Semantc segmentaton results 732

6 39 8 Vol.39 No August mean IU mean IU CRF CRF Ctyscapes SegModel mean IU IU IU CRF CRF 4 Caffe [1] Long J, Shelhamer E, Darrell T. Fully convolutonal networks for semantc segmentaton[c]//ieee Conference on Computer Vson and Pattern Recognton, 2015: [2] Zheng S, Jayasumana S, Romeraparedes B, et al. Condtonal random felds as recurrent neural networks[c]//ieee Conference on Computer Vson and Pattern Recognton, 2015: [3] Chen L C, Papandreou G, Kokknos I, et al. Semantc mage segmentaton wth deep convolutonal nets and fully connected CRFs[J]. Computer Scence, 2014(4): [4] Noh H, Hong S, Han B. Learnng deconvoluton network for semantc segmentaton[c]//proceedngs of the IEEE Internatonal Conference on Computer Vson, 2015: [5] Pzer S M, Amburn E P, Austn J D, et al. Adaptve Hstogram equalzaton and ts varatons[j]. Computer Vson, Graphcs, and Image Processng, 1987, 39(3): [6] Smonyan K, Zsserman A. Very deep convolutonal networks for large-scale mage recognton[2014] [DB/OL]. arxv preprnt arxv: [7] Krähenbühl P, Koltun V. Effcent Inference n fully connected CRFs wth Gaussan edge potentals[c]//advances n Neural Informaton Processng Systems, 2012: [8] Cordts M, Omran M, Ramos S, et al. The ctyscapes dataset for semantc urban scene understandng[c]//proceedngs of the IEEE Conference on Computer Vson and Pattern Recognton, 2016:

Microsoft Word - 系统建设1.doc

Microsoft Word - 系统建设1.doc 计算机系统应用 http://www.c-s-a.org.cn 20 年第 20 卷第 4 期一种高性能近红外光人脸检测和眼睛定位算法张昌明, 童卫青, 王燕群 ( 华东师范大学计算机科学技术系, 上海 20024) 摘要 : 提出了一种新型的近红外光人脸检测和眼睛