現今，有很多人因眼睛疾病而影響視力，其中以黃斑病變(AMD)以及視網膜退化病變(RP)這兩種視網膜疾病更屬常見。而這兩種視網膜疾病嚴重會造成眼睛失明，有很多療法能幫助病患恢復視力，其中包括用以取代外視網膜神經的人工視網膜為常見方法。
除了讓人工視網膜晶片能取代受損視網膜細胞的感光功能來達到建構影像以外，也要達到人類眼睛對視覺的高對比敏感能力。利用影像處理的方式，先將訊號經過處理加深輪廓後再送往腦神經以增加人工視網膜晶片的感光能力。所以將常應用於輪廓偵測或動態偵測等的拉普拉斯運算，這種二階微分的演算法用來模擬人眼的側抑制能力，並用脈衝邏輯運算的方式來實現，進而使晶片達到更接近人眼的影像。
本篇會更著重於模擬錯誤注入對於系統輸出的影響，除了探討多種數量錯誤對影像的影響，也會研究錯誤被放大的主因。另外，用PSNR的方法輔助檢測影像的破懷程度，進而得到一些方法來診斷錯誤對最後輸出的影響範圍，這樣可以幫助以後對受損晶片快速測試且診斷出錯誤成因。最後，也會針對找出最適合模擬人眼的拉普拉斯運算子來做各種比較分析，不管正規的拉普拉斯運算子還是非正規拉普拉斯運算子都會進行探討。

Nowadays, many people have the visual impairment due to eye diseases. There are two common retinal diseases, one is macular degeneration (AMD) and the other one is retinal degeneration (RP). Because these retinal diseases can make patients blind totally, there are many treatments to help patients restore their vision, including the artificial retinal prosthesis that can replace the function of outer retinal nerves.
The retinal prosthesis not only should replace the function of light detection from optical nerves, but also make the implanters’ vision similar with human eyes which have the high contrast sensitivity. Through the method of image processing, the signal is processed by contrast enhancement before an image sent to the brain such that the visual ability increases. Therefore, Laplacian operator, which is used to detect edges or motivations of images, is a second derivative for 2D image to simulate the lateral inhibition. Then, using pulse logic approach to implement Laplacian operator makes the output vison similar with human vision.
In this thesis, it focuses on analysis for the fault injections in the image. It discusses not only many cases with different amount of defective pixels but also the reason why the fault is amplified. In addition, use an objective image quality measure, PSNR to compare the fault injected image with original image. Then, there are some conclusions to help us test chip rapidly and diagnose which kind of fault through the output easily. Finally, it shows the comparison among different Laplacian operators and choose the best one to simulate visual ability of human eyes. Not only normal Laplacian operator but also abnormal Laplacian operator is discussed in this thesis.

中文摘要............................................................i
Abstract...........................................................ii
誌謝...............................................................iii
Table of Content...................................................iv
List of Figures....................................................vi
List of Tables.....................................................xii
Ch1. Introduction..................................................1
1.1 Background.....................................................1
1.2 Motivation.....................................................2
1.3 Lateral Inhibition.............................................3
1.4 Imitate Lateral Inhibition by Laplacian Operator...............5
1.5 Organization of Thesis.........................................7
Ch2. Retinal Prosthesis Chip.......................................9
2.1 Related Work...................................................9
2.2 Light Intensity Detection......................................10
2.3 Pulse Dependent Voltage Generation.............................11
2.4 Converting Pulses to Electrical Signals........................12
2.5 Pulse Logic Based Retinal Prosthesis Signal Processing.........13
Ch3. Laplacian Operator and Faults on Chip.........................17
3.1 Mathematical Definition of Laplacian Operator..................17
3.2 Pixel Arrangement..............................................20
3.3 Laplacian Operation in Our Chip................................22
3.4 Physical Definition of Laplacian Operator......................24
3.5 Fault Generation in The Chip...................................28
3.6 Image Quality Measure..........................................30
Ch4. Fault Injections on Pixel Block...............................32
4.1 Fault Injection Flow...........................................32
4.2 Analysis on One-Pixel Fault....................................34
4.3 Analysis on Many-Pixels Fault..................................46
4.4 Global Variations..............................................53
4.5 Local Variations...............................................58
Ch 5. Different Weights of Laplacian Operator......................69
5.1 Four Normal Laplacian Operators................................69
5.2 Abnormal Laplacian Operator....................................77
5.3 Change Number of Light Intensity Level.........................88
Ch 6. Conclusion and Future Work...................................94
6.1 Conclusion.....................................................94
6.2 Future Work....................................................95
Reference..........................................................97

[1] J. D. Weiland, W. Liu, and M. S. Humayun, “Retinal Prosthesis,” Annual Review of Biomedical Engineering, vol. 7, pp.361-401, 2005.
[2] GBD 2015 Disease and Injury Incidence and Prevalence Collaborators, “Global, regional, and national incidence, prevalence, and years lived with disability for 310 diseases and injuries, 1990-2015: a systematic analysis for the Global Burden of Disease Study 2015,” Global Burden of Disease Study, The Lancet. 388 (10053), Oct. 2016.
[3] A. Rothermel, L. Liu, N. P. Aryan, M. Fischer, J. Wuenschmann, S. Kibbel, and A. Harscher, “A CMOS Chip With Active Pixel Array and Specific Test Features for Subretinal Implantation”, IEEE Journal of Solid-State circuits, vol. 44, no. 1, Jan. 2009.
[4] C.-L. Lee and C.-C. Hsieh, “A 0.8-V 4096-Pixels CMOS Sense-and-Stimulus Imager for Retinal Prosthesis,” IEEE Transactions on Electron Devices, vol. 60, no. 3, pp. 1162-1168, Mar. 2013.
[5] A.C. Weitz, “Interphase gap as a means to reduce electrical stimulation thresholds for epiretinal prostheses,” Journal of neural engineering, 11.1: 016007, 2014.
[6] Z.-H. Ye, “Laplacian Algorithm Applied to Retinal Prosthesis Chip,” MS thesis, National Tsing Hua University, Aug. 2016.
[7] Hoffman JD; Frankel S, “Numerical methods for engineers and scientists,” CRC Press, Boca Raton, P264-P267, 2001.
[8] B. Fornberg and N. Flyer, “A Primer on Radial Basis Functions with Applications to the Geosciences”, SIAM-Society for Industrial and Applied Mathematics, 2015
[9] L. Velho, A. C. Frery, and J. Gomes, “Image Processing for Computer Graphics and Vision,” 2nd edn. Springer, 2009
[10] C.-W. Hsu, “Retinal Prosthesis Chip Output Analysis and Peripheral Circuits,” MS thesis, National Tsing Hua University, July 2019.
[11] Y. A. Y. Al-Najjar, and Der C. Soong, “Comparison of Image Quality Assessment: PSNR, HVS, SSIM, UIQI,” International Journal of Scientific & Engineering Research, vol. 3, no. 8, pp. 1-5, Aug. 2012.
[12] D.-S. Park, J.-H. Kim, H.-S. Kim, J.-H. Park, J.-K. Shin, and M. Lee, “A foveated-structure CMOS retina chip for edge detection with local light adaptation,”, Sensors and Actuators A Physical 108, pp. 75-80, Nov. 2003.
[13] H. Naganuma, K. Kiyoyama, and T. Tanaka, “A 37 × 37 Pixels Artificial Retina Chip with Edge Enhancement Function for 3-D Stacked Fully Implantable Retinal Prosthesis,” 2012 IEEE Biomedical Circuits and Systems Conference, 2012.
[14] K. Shimokawa et al., “Experimental evaluation of stimulus current generator with Laplacian edge-enhancement for 3-D stacked retinal prosthesis chip,” 2017 IEEE Biomedical Circuits and Systems Conference (BioCAS), Turin, pp. 1-4, 2017.
[15] Z. Wang, A. C. Bovik, H. R. Sheikh, and E. P. Simoncelli, “Image Quality Assessment: From Error Visibility to Structural Similarity,” IEEE Transactions On Image Processing, Vol. 13, no. 4, pp. 600-612, Apr. 2004.
[16] J. Chen, T.-S. Chen, K.-J. Ma and P.-H. Wang, “Digital Watermarking Based on Neural Network Technology for Grayscale Images,” Encyclopedia of Multimedia Technology and Networking. IGI Global, pp.204-212, 2005. Web. 23 Nov. 2019.
[17] P. Geetha, “Survey of Medical Image Compression Techniques and Comparative Analysis,” Research Developments in Computer Vision and Image Processing: Methodologies and Applications. IGI Global, pp.327-356, 2014.Web. 23 Nov. 2019.
[18] N. Tran et al., “A Complete 256-Electrode Retinal Prosthesis Chip,” in IEEE Journal of Solid-State Circuits, vol. 49, no. 3, pp. 751-765, March 2014.
[19] N. Sharmili, V. Bhujanga Rao, P. Seetharamaiah, and N. Swapna, “A prototype 1024 electrode embedded computer based epiretinal prosthesis system,” in IEEE 3rd International Conference on Signal Processing and Integrated Networks, pp. 337-341, 2016.
[20] A. Soltan et al., “High Density, High RadianceμLED Matrix for Optogenetic Retinal Prostheses and Planar Neural Stimulation,” in IEEE Transactions on Biomedical Circuits and Systems, vol. 11, no. 2, pp. 347-359, April 2017.
[21] I. A. Mashhadi, M. Pahlevani, S. Hor, H. Pahlevani and E. Adib, “A New Wireless Power-Transfer Circuit for Retinal Prosthesis,” in IEEE Transactions on Power Electronics, vol. 34, no. 7, pp. 6425-6439, July 2019.
[22] S. Oh et al., “Light-Controlled Biphasic Current Stimulator IC Using CMOS Image Sensors for High-Resolution Retinal Prosthesis and In Vitro Experimental Results With rd1 Mouse,” in IEEE Transactions on Biomedical Engineering, vol. 62, no. 1, pp. 70-79, Jan. 2015.
[23] C.-Y. Wu, et al., “A CMOS 256-pixel Photovoltaics-powered Implantable Chip with Active Pixel Sensors and Iridium-oxide Electrodes for Subretinal Prostheses,” Sensors and Materials., vol. 30, no. 2, pp. 193-211, Jan. 2018.
[24] O. Wong, P. Chen and C. Wu, “A fully-integrated charge pump for self-powered implantable retinal prostheses,” 2016 IEEE International Symposium on Radio-Frequency Integration Technology (RFIT), Taipei, pp. 1-3, Sep. 2016.
[25] P. Kuo et al., “Improved Charge Pump Design and Ex Vivo Experimental Validation of CMOS 256-Pixel Photovoltaic-Powered Subretinal Prosthetic Chip,” in IEEE Transactions on Biomedical Engineering, Sep. 2019.
[26] C. D. Eiber, S. Dokos, N. H. Lovell and G. J. Suaning, “Multipolar Field Shaping in a Suprachoroidal Visual Prosthesis,” in IEEE Transactions on Neural Systems and Rehabilitation Engineering, vol. 25, no. 12, pp. 2480-2487, Dec. 2017.
[27] T. Noda, H. Takehara, K. Sasagawa, T. Tokuda and J. Ohta, “Neural stimulators for retinal prosthesis embedded with CMOS microchips,” 2016 IEEE 59th International Midwest Symposium on Circuits and Systems (MWSCAS), Abu Dhabi, pp. 1-4, 2016.
[28] Z. Zheng, M. Chen, J. Zhou and G. Wang, “A CMOS Temperature Sensor With Single-Point Calibration for Retinal Prosthesis,” 2018 IEEE Asia Pacific Conference on Circuits and Systems (APCCAS), Chengdu, pp. 147-150, 2018.
[29] H. Song, G. Kim and H. Ko, “Triple cascode high output impedance current stimulator with dynamic element matching for retinal prostheses,” 2015 15th International Conference on Control, Automation and Systems (ICCAS), Busan, pp. 1866-1869, 2015.
[30] R. A. Zawawi, J. Kim, J. Park, S. Kim, A. A. Manaf and J. Kim, “A 3.3 V, 8.89 μA and 5.5 ppm/°C CMOS bandgap voltage reference for power telemetry in retinal prosthesis systems,” 2018 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), Honolulu, HI, pp. 2977-2980, 2018.