Signal-to-noise ratio of a linear camera

import%20marimo%0A%0A__generated_with%20%3D%20%220.25.1%22%0Aapp%20%3D%20marimo.App(app_title%3D%22Signal-to-noise%20ratio%20of%20a%20linear%20camera%22)%0A%0A%0A%40app.cell(hide_code%3DTrue)%0Adef%20_(mo)%3A%0A%20%20%20%20mo.md(r%22%22%22%0A%20%20%20%20%23%20Signal-to-noise%20ratio%20of%20a%20linear%20camera%0A%0A%20%20%20%20The%20signal-to-noise%20ratio%20(SNR)%20tells%20how%20well%20a%20pixel's%20mean%20signal%20stands%20out%0A%20%20%20%20from%20its%20temporal%20fluctuations.%20This%20page%20follows%20the%20linear%20camera%20model%20of%0A%20%20%20%20the%20EMVA%201288%20standard%20(EMVA%201288%2C%202021)%2C%20%C2%A72.4%20and%20%C2%A72.6.%0A%0A%20%20%20%20%23%23%20Model%0A%0A%20%20%20%20A%20pixel%20collects%20on%20average%20%24%5Cmu_p%24%20photons%20during%20the%20exposure.%20A%20fraction%0A%20%20%20%20%24%5Ceta%24%20(the%20quantum%20efficiency%2C%20in%20e%2Fph)%20is%20converted%20into%20electrons%2C%20so%20the%0A%20%20%20%20mean%20signal%20is%20%24%5Cmu_e%20%3D%20%5Ceta%5Cmu_p%24.%20Three%20independent%20noise%20sources%20add%20in%0A%20%20%20%20variance%3A%0A%0A%20%20%20%20-%20**Photon%20shot%20noise.**%20Photon%20arrival%20is%20a%20Poisson%20process%2C%20so%20the%20variance%0A%20%20%20%20%20%20in%20electrons%20equals%20the%20mean%3A%20%24%5Csigma_e%5E2%20%3D%20%5Cmu_e%24.%0A%20%20%20%20-%20**Read%20noise**%20%24%5Csigma_%7Bread%7D%24%2C%20in%20electrons.%20It%20does%20not%20depend%20on%20the%20signal%0A%20%20%20%20%20%20and%20lumps%20together%20everything%20in%20the%20readout%20chain.%0A%20%20%20%20-%20**Quantization%20noise.**%20The%20ADC%20rounds%20to%20whole%20digital%20numbers.%20For%20a%0A%20%20%20%20%20%20uniform%20rounding%20error%20the%20variance%20is%20%241%2F12%5C%20%5Cmathrm%7BDN%7D%5E2%24.%20Dividing%20by%20the%0A%20%20%20%20%20%20squared%20system%20gain%20%24K%5E2%24%20(in%20DN%2Fe)%20refers%20it%20back%20to%20electrons.%0A%0A%20%20%20%20Combining%20read%20and%20quantization%20noise%20into%20one%20dark%20noise%20term%20gives%0A%0A%20%20%20%20%24%24%0A%20%20%20%20%5Csigma_d%5E2%20%3D%20%5Csigma_%7Bread%7D%5E2%20%2B%20%5Cfrac%7B1%2F12%7D%7BK%5E2%7D%2C%0A%20%20%20%20%5Cqquad%0A%20%20%20%20%5Cmathrm%7BSNR%7D%20%3D%20%5Cfrac%7B%5Ceta%5C%2C%5Cmu_p%7D%7B%5Csqrt%7B%5Csigma_d%5E2%20%2B%20%5Ceta%5C%2C%5Cmu_p%7D%7D.%0A%20%20%20%20%24%24%0A%0A%20%20%20%20EMVA%201288%20writes%20the%20quantization%20term%20separately%2C%20as%20%24%5Csigma_q%5E2%2FK%5E2%24%20next%20to%0A%20%20%20%20%24%5Csigma_d%5E2%24.%20The%20result%20is%20the%20same.%20In%20code%3A%20%60photonlab.noise.snr%60%20and%0A%20%20%20%20%60photonlab.noise.dark_variance%60.%0A%0A%20%20%20%20%23%23%20Two%20regimes%0A%0A%20%20%20%20-%20**Low%20light**%20(%24%5Ceta%5Cmu_p%20%5Cll%20%5Csigma_d%5E2%24)%3A%20%24%5Cmathrm%7BSNR%7D%20%5Capprox%20%5Ceta%5Cmu_p%2F%5Csigma_d%24.%0A%20%20%20%20%20%20The%20SNR%20grows%20linearly%20with%20exposure%2C%20and%20lowering%20the%20read%20noise%20pays%20off%20directly.%0A%20%20%20%20-%20**High%20light**%20(%24%5Ceta%5Cmu_p%20%5Cgg%20%5Csigma_d%5E2%24)%3A%20%24%5Cmathrm%7BSNR%7D%20%5Capprox%20%5Csqrt%7B%5Ceta%5Cmu_p%7D%24.%0A%20%20%20%20%20%20Shot%20noise%20dominates%2C%20and%20only%20collecting%20more%20photons%20helps.%20The%20maximum%20SNR%0A%20%20%20%20%20%20is%20reached%20at%20saturation%2C%20%24%5Csqrt%7B%5Cmu_%7Be%2C%5Cmathrm%7Bsat%7D%7D%7D%24.%0A%0A%20%20%20%20On%20a%20log-log%20plot%20this%20gives%20a%20slope%20of%201%20at%20low%20light%2C%20bending%20to%20a%20slope%20of%0A%20%20%20%201%2F2.%20The%20photon%20transfer%20method%20builds%20on%20the%20same%20behavior%20(Janesick%2C%202007).%0A%0A%20%20%20%20%23%23%20Derived%20figures%0A%0A%20%20%20%20-%20**Absolute%20sensitivity%20threshold**%20%24%5Cmu_%7Bp%2C%5Cmin%7D%24%3A%20the%20photon%20count%20where%0A%20%20%20%20%20%20SNR%20%3D%201.%20Setting%20SNR%20%3D%201%20and%20solving%20the%20quadratic%20gives%0A%20%20%20%20%20%20%24%5Cmu_%7Bp%2C%5Cmin%7D%20%3D%20%5Cfrac%7B1%7D%7B%5Ceta%7D%5Cleft(%5Cfrac%7B1%7D%7B2%7D%20%2B%20%5Csqrt%7B%5Cfrac%7B1%7D%7B4%7D%20%2B%20%5Csigma_d%5E2%7D%5Cright)%24%0A%20%20%20%20%20%20(EMVA%201288%20%C2%A72.7).%20In%20code%3A%20%60photonlab.sensor.absolute_sensitivity_threshold%60.%0A%20%20%20%20-%20**Dynamic%20range**%3A%20%24%5Cmathrm%7BDR%7D%20%3D%20%5Cmu_%7Bp%2C%5Cmathrm%7Bsat%7D%7D%2F%5Cmu_%7Bp%2C%5Cmin%7D%24%2C%20quoted%0A%20%20%20%20%20%20in%20dB%20as%20%2420%5Clog_%7B10%7D%5Cmathrm%7BDR%7D%24%20(EMVA%201288%20%C2%A72.7%20and%20%C2%A76.6).%20In%20code%3A%0A%20%20%20%20%20%20%60photonlab.sensor.dynamic_range%60.%0A%0A%20%20%20%20%23%23%20Try%20it%0A%0A%20%20%20%20Pick%20a%20sensor%20from%20the%20catalog%20(%60photonlab.catalog%60)%20to%20load%20its%20values%2C%20then%20move%0A%20%20%20%20the%20sliders.%20In%20this%20example%20the%20gain%20%24K%24%20is%20chosen%20so%20that%20the%20full%20well%20lands%0A%20%20%20%20exactly%20on%20the%20highest%20ADC%20code%20(%60photonlab.sensor.gain_for_full_scale%60).%0A%20%20%20%20%22%22%22)%0A%20%20%20%20return%0A%0A%0A%40app.cell%0Adef%20_(catalog%2C%20mo)%3A%0A%20%20%20%20sensor_choice%20%3D%20mo.ui.dropdown(%0A%20%20%20%20%20%20%20%20%7Bcatalog.sensor(i).name%3A%20i%20for%20i%20in%20catalog.sensors()%7D%2C%0A%20%20%20%20%20%20%20%20value%3Dcatalog.sensor(catalog.sensors()%5B0%5D).name%2C%0A%20%20%20%20%20%20%20%20label%3D%22Sensor%22%2C%0A%20%20%20%20)%0A%20%20%20%20sensor_choice%0A%20%20%20%20return%20(sensor_choice%2C)%0A%0A%0A%40app.cell%0Adef%20_(catalog%2C%20mo%2C%20sensor_choice)%3A%0A%20%20%20%20sensor%20%3D%20catalog.sensor(sensor_choice.value)%0A%0A%0A%20%20%20%20def%20_start(value%2C%20low%2C%20high)%3A%0A%20%20%20%20%20%20%20%20return%20min(max(value%2C%20low)%2C%20high)%20%20%23%20keep%20the%20sensor%20value%20inside%20the%20slider%20range%0A%0A%0A%20%20%20%20qe%20%3D%20mo.ui.slider(0.05%2C%201.0%2C%20step%3D0.01%2C%20value%3D_start(sensor.qe.magnitude%2C%200.05%2C%201.0)%2C%0A%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20show_value%3DTrue%2C%20label%3D%22Quantum%20efficiency%20%CE%B7%20%5Be%2Fph%5D%22)%0A%20%20%20%20read_noise%20%3D%20mo.ui.slider(0.5%2C%2020.0%2C%20step%3D0.1%2C%20value%3D_start(sensor.read_noise.magnitude%2C%200.5%2C%2020.0)%2C%0A%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20show_value%3DTrue%2C%20label%3D%22Read%20noise%20%CF%83_read%20%5Be%5D%22)%0A%20%20%20%20full_well%20%3D%20mo.ui.slider(1_000%2C%20100_000%2C%20step%3D500%2C%20value%3D_start(sensor.full_well.magnitude%2C%201_000%2C%20100_000)%2C%0A%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20show_value%3DTrue%2C%20label%3D%22Full%20well%20%5Be%5D%22)%0A%20%20%20%20adc_bits%20%3D%20mo.ui.slider(8%2C%2016%2C%20step%3D1%2C%20value%3D_start(sensor.adc_bits%2C%208%2C%2016)%2C%0A%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20%20show_value%3DTrue%2C%20label%3D%22ADC%20resolution%20%5Bbits%5D%22)%0A%20%20%20%20mo.vstack(%5B%0A%20%20%20%20%20%20%20%20mo.md(f%22*%7Bsensor.description%7D*%20Values%3A%20%7Bsensor.kind%7D.%22)%2C%0A%20%20%20%20%20%20%20%20qe%2C%20read_noise%2C%20full_well%2C%20adc_bits%2C%0A%20%20%20%20%5D)%0A%20%20%20%20return%20adc_bits%2C%20full_well%2C%20qe%2C%20read_noise%0A%0A%0A%40app.cell%0Adef%20_(%0A%20%20%20%20absolute_sensitivity_threshold%2C%0A%20%20%20%20adc_bits%2C%0A%20%20%20%20dynamic_range%2C%0A%20%20%20%20full_well%2C%0A%20%20%20%20gain_for_full_scale%2C%0A%20%20%20%20nb%2C%0A%20%20%20%20qe%2C%0A%20%20%20%20read_noise%2C%0A%20%20%20%20snr%2C%0A%20%20%20%20ureg%2C%0A)%3A%0A%20%20%20%20eta%20%3D%20qe.value%20*%20ureg(%22e%2Fph%22)%0A%20%20%20%20sigma_read%20%3D%20read_noise.value%20*%20ureg.e%0A%20%20%20%20K%20%3D%20gain_for_full_scale(full_well.value%20*%20ureg.e%2C%20adc_bits.value%20*%20ureg.dimensionless)%0A%20%20%20%20mu_p_sat%20%3D%20(full_well.value%20*%20ureg.e%20%2F%20eta).to(%22ph%22)%0A%20%20%20%20mu_p_min%20%3D%20absolute_sensitivity_threshold(eta%2C%20sigma_read%2C%20K)%0A%0A%20%20%20%20nb.table(%5B%0A%20%20%20%20%20%20%20%20(%22System%20gain%20K%22%2C%20K)%2C%0A%20%20%20%20%20%20%20%20(%22Saturation%20%CE%BCp%2Csat%22%2C%20mu_p_sat%2C%20%22%7B%3A%2C.0f%7D%22)%2C%0A%20%20%20%20%20%20%20%20(%22Sensitivity%20threshold%20%CE%BCp%2Cmin%20(SNR%20%3D%201)%22%2C%20mu_p_min%2C%20%22%7B%3A.3g%7D%22)%2C%0A%20%20%20%20%20%20%20%20(%22Dynamic%20range%22%2C%20dynamic_range(mu_p_sat%2C%20eta%2C%20sigma_read%2C%20K)%2C%20%22%7B%3A.1f%7D%22)%2C%0A%20%20%20%20%20%20%20%20(%22Maximum%20SNR%22%2C%20snr(mu_p_sat%2C%20eta%2C%20sigma_read%2C%20K)%2C%20%22%7B%3A.1f%7D%22)%2C%0A%20%20%20%20%5D)%0A%20%20%20%20return%20K%2C%20eta%2C%20mu_p_min%2C%20mu_p_sat%2C%20sigma_read%0A%0A%0A%40app.cell%0Adef%20_(K%2C%20eta%2C%20mu_p_min%2C%20mu_p_sat%2C%20nb%2C%20np%2C%20sigma_read%2C%20snr%2C%20ureg)%3A%0A%20%20%20%20photons%20%3D%20np.geomspace(max(mu_p_min.magnitude%20%2F%2010%2C%200.1)%2C%20mu_p_sat.magnitude%2C%20300)%20*%20ureg.ph%0A%20%20%20%20nb.loglog_plot(%0A%20%20%20%20%20%20%20%20photons%2C%20snr(photons%2C%20eta%2C%20sigma_read%2C%20K)%2C%0A%20%20%20%20%20%20%20%20xlabel%3D%22Photons%20per%20pixel%20%CE%BCp%22%2C%20ylabel%3D%22SNR%22%2C%20label%3D%22Linear%20model%22%2C%0A%20%20%20%20%20%20%20%20vlines%3D%7B%22%CE%BCp%2Cmin%20(SNR%20%3D%201)%22%3A%20mu_p_min%2C%20%22%CE%BCp%2Csat%22%3A%20mu_p_sat%7D%2C%0A%20%20%20%20%20%20%20%20hlines%3D%7B%22SNR%20%3D%201%22%3A%201%20*%20ureg.dimensionless%7D%2C%0A%20%20%20%20)%0A%20%20%20%20return%0A%0A%0A%40app.cell(hide_code%3DTrue)%0Adef%20_(mo)%3A%0A%20%20%20%20mo.md(r%22%22%22%0A%20%20%20%20%23%23%20Limits%20of%20the%20model%0A%0A%20%20%20%20-%20Linear%20sensor%20below%20saturation%3B%20no%20dark%20current%2C%20so%20short%20exposures.%0A%20%20%20%20-%20Temporal%20noise%20of%20one%20pixel%20only.%20Spatial%20non-uniformity%20(DSNU%2C%20PRNU)%20is%20ignored.%0A%20%20%20%20-%20The%201%2F12%20DN%C2%B2%20quantization%20variance%20assumes%20the%20noise%20spans%20more%20than%20about%20one%20DN.%0A%20%20%20%20-%20Saturation%20capacity%20is%20not%20the%20same%20as%20full-well%20capacity.%20The%20ADC%20often%20clips%0A%20%20%20%20%20%20first.%0A%20%20%20%20%22%22%22)%0A%20%20%20%20return%0A%0A%0A%40app.cell%0Adef%20_(nb%2C%20refs)%3A%0A%20%20%20%20nb.references_footer(refs%2C%20%22emva1288%3A2021%2C%20janesick%3A2007%22)%0A%20%20%20%20return%0A%0A%0A%40app.cell(hide_code%3DTrue)%0Adef%20_(photonlab_loaded)%3A%0A%20%20%20%20%23%20Functions%20used%20on%20this%20page.%20Add%20imports%20here.%0A%20%20%20%20photonlab_loaded%20%20%23%20runs%20after%20SETUP%20below%2C%20which%20loads%20photonlab%20in%20the%20browser%0A%20%20%20%20import%20matplotlib.pyplot%20%20%23%20noqa%3A%20F401%20%20(tells%20the%20browser%20to%20load%20matplotlib%20for%20nb.loglog_plot)%0A%20%20%20%20import%20numpy%20as%20np%0A%0A%20%20%20%20from%20photonlab%20import%20catalog%2C%20nb%0A%20%20%20%20from%20photonlab.noise%20import%20snr%0A%20%20%20%20from%20photonlab.sensor%20import%20absolute_sensitivity_threshold%2C%20dynamic_range%2C%20gain_for_full_scale%0A%20%20%20%20from%20photonlab.units%20import%20ureg%0A%0A%20%20%20%20return%20(%0A%20%20%20%20%20%20%20%20absolute_sensitivity_threshold%2C%0A%20%20%20%20%20%20%20%20catalog%2C%0A%20%20%20%20%20%20%20%20dynamic_range%2C%0A%20%20%20%20%20%20%20%20gain_for_full_scale%2C%0A%20%20%20%20%20%20%20%20nb%2C%0A%20%20%20%20%20%20%20%20np%2C%0A%20%20%20%20%20%20%20%20snr%2C%0A%20%20%20%20%20%20%20%20ureg%2C%0A%20%20%20%20)%0A%0A%0A%40app.cell(hide_code%3DTrue)%0Aasync%20def%20_()%3A%0A%20%20%20%20%23%20SETUP%20%E2%80%94%20do%20not%20edit.%20Loads%20photonlab%20and%20the%20reference%20texts%2C%20locally%20and%20in%20the%20browser.%0A%20%20%20%20import%20base64%0A%20%20%20%20import%20sys%0A%20%20%20%20from%20pathlib%20import%20Path%0A%0A%20%20%20%20import%20marimo%20as%20mo%0A%0A%20%20%20%20import%20json%0A%20%20%20%20EMBEDDED%20%3D%20json.loads('%7B%22wheel%22%3A%20%22photonlab-0.1.0-py3-none-any.whl%22%2C%20%22wheel_b64%22%3A%20%22UEsDBBQAAAgAAAAAIQAAAAAAAAAAAAAAAAAKAAAAcGhvdG9ubGFiL1BLAwQUAAAICAAAACEAG3hheMUAAAA%2FAQAAFQAAAHBob3RvbmxhYi9fX2luaXRfXy5weV2QTU7EMAxG9zmF5RVI0ANUYsERWCPUcVNPa%2BE4UeLwc3s0pQOIrb%2F3%2FFlGxLJlz6Y0jyCJVrH1vkVhiwyRNHYll2wN3sU34I%2BiEsWhm3gbQnhUhdJnlQjnbvEbdXplIFugsvdqUMQcnjqZiws3ONecwDeGtlHlJVRepXn9BDEYU17G089Rw150GgIihrCLkkqurjIPiZ0WcjpG8Ma1SbaD%2B7fjCvXKawjTdMDTBA9X8eb3GXh7YUh1z5%2FxYuEd4B8PX8IXUEsDBBQAAAgAAAAAIQAAAAAAAAAAAAAAAAASAAAAcGhvdG9ubGFiL2NhdGFsb2cvUEsDBBQAAAgIAAAAIQDLVDMCrQUAACwOAAAdAAAAcGhvdG9ubGFiL2NhdGFsb2cvX19pbml0X18ucHmtV21v2zYQ%2Fq5fcWM%2FVCoUNcW%2BDC48tMAKrGjXbWg2YDACi5ZOFhuKVEgqjpfmvw9HUi9uirbY5g%2B2RR7vuXvujWKMvUdltQGuapCoLBistKltDtZpgzVwC3%2B9%2FOUtKLx14DS4VlhohMRwRPM6CP0%2BcOWEE2iLJPlVYVQEPZogLhSUpfVo9mlZgjZQlgSJ9PgcXItBUPEOQVi%2FIOoi%2BZPLAS1wg2CdEWpv4SBcC4MSzkJalt8X5zB0ZZlDWT47Pz8%2FByzL7DlwOBit9l4QGi6kBa28yUXym9E3qLiqPBQZGexdkZF6MBWWJaQcDDa22IkdXOEx8y6X5ZVQdVkmac0dty2ig4%2FQIbcD8fURuLVDh3VWwEV0yYJtRQ9CWVGj9%2BvQIsocLPGJUHHHpd4nXFoNB22uLLHFwbacVP588ctb6PkeiyQBAGiM7qBvtdNK8h2IrtfGTUpIwsJ6fC4C5SnDW971Es%2FC8xlnWRAtmkHK7QEl2VNcI30fRO3ahDGWJB6MPK0ktxbtCDctBYmwKsWuMBj4myQ9A0kSn468k1Hr5EIRQhklBoP7JHnz%2Bt1P72ENKZtoZjmwkWf6H4lmWZIkLyZ70sbov1GtL8yAWeKXICT5yjss6hXlkf9PqTY%2F9eIW5bYXrmpXoHcfsHIAj2AzdJd%2BnzyTgxNarcANvUTaTT1XObQo9q3LKHBej%2FVHeF1td8LZFQjlQvBGtk8gcETg9VZpYfFzu9cnqz3yKyoipTuhuIQNPu3bUXBbDeYGRzM%2FwjutENbhxxvNb1Ci2rsWNqq7zOEao4IMuDH8GMyv0VZG9MFl6wysgbGQNj7GYfFEvd%2BlCpkOjGEKyfuiN7pH444RoAFPYGpRNlmIUGDCDUaBXy4WgYEn4JfmWGzOL7PC6ZR1HWXC5zFCdP4DyLMTkK%2Bk21tU9mvJ1uiKy22IwUmwu5htzVYN3Q7NChqpeUge0fE9bithKjnnwhzW6ez%2FHLfkEbziVQt9e7Si4hIagbKmxllpdYPGYU2jgWqYKnnuvLXoUFmhFRguqHlM%2FfePd68vqLzv2IJ3tgI2dFTaU5XQEtLKXBrT0nX4%2B7RvWT7FFNiSWRLovMYldXH1PkkSSo6tqG3aaFmjickRE8Nqci3tCwpdYbDTN2iHphG3KSuokbEMGm2AWntoc%2Bl22%2FPqiqC2WfFBC9Vz147KC%2BHQ1MKkofWefkQDEQhVbWnCTSDZaChxF5XlUEmbx6m1FXW0nOBg%2FQ3G5NCwu%2Bn0fUQKWdaA0s6rKoTdkq50WTUUSniDx1fGaJM2TGm4C0o3q7Nnl%2FfweKH48XO4UvqgVnC35Pk%2BYhl%2BgLUfCoXlDQYPPbIPuMNbl6KqdC3Ufs0G15z9QHxEMw0%2FFHt0KRM1y%2BC79UzHp%2Bb6S8RocLT2%2FumnBKxA1NAN1gFeD1ye3klGi0NubHtuLKYhCPwwheiT5WAHY%2BzleCGqReXowkT%2BhVHu71dnP8YpRT2dGki45CyuVTSOZ8pITeqBg1HRD1ifkvKAKSrxk%2BHpYy0U%2BIH7Zdpmtla%2BVQSqdgjURnQDd17HGFqqDN8o8nADEwp81VMVdHaZUaIZOwo1isO8Ed3d%2BN1LWPu7QRE5OabWmXTeznyHJqjJazZOQvZAtV8ObG1msdA%2BR9jFOt1ETsw6tWRz%2BniTFd2W25SpLraIGzLA69qwefiyy8scvFD%2BL5T7zvdQ%2FTUGtWF7VjyzMs%2B1h7x4vxcC5Lm%2FRaQPdk7qoZI2ffKEEpIm0aB80VMud8JaofYhwjbm1cWxD2kVyyZeUsMPlS8NoWwuiqmM3mpe%2B3QLomNR7Y70qjBVSCxR30xYfOdgedSVx7PUNCM8vYak9HUCPY9xtgBevCV9ETa82rDcq8n9sQViNCr1QFJYt7HOXE5or2tLBcWlPHV00QUiGPXUycWlP%2FgtypevfJ9XHd3Ikn8AUEsDBBQAAAgAAAAAIQAAAAAAAAAAAAAAAAAZAAAAcGhvdG9ubGFiL2NhdGFsb2cvbGVuc2VzL1BLAwQUAAAICAAAACEAEEIwF4kAAACrAAAALAAAAHBob3RvbmxhYi9jYXRhbG9nL2xlbnNlcy9leGFtcGxlLWxlbnMtYS55YW1sJY3BCsIwEAXv%2BYoHnluoB5F48qAn8RfKmmxrYDcp2UT8fCleh2HmgHsKLZVMAuFsWEpFLMFAOaKxNRvxLA2EyiTYaok9tAtIBB%2BSzgYy68pxdCl68Jd0Ex722EAuk7LH7Q%2Fx2A9Xt5RAMgvntb09phNU3TLnri%2BuHsfx7JLSynNINQh7TNNu%2FABQSwMEFAAACAgAAAAhAAvPgQZ8AAAAlgAAACwAAABwaG90b25sYWIvY2F0YWxvZy9sZW5zZXMvZXhhbXBsZS1sZW5zLWIueWFtbCXNMQoCMRBG4T6n%2BMF6A3ZL7AStxCsss8lEF2aSJZOIx5fF9hXfO%2BG%2Bxb7VQgLhYsi1IdVooJLQ2bp5PGsHoTEJ9lbTiP0CEsGHZLCBzIZy8m5LAfwl3YWnA5tWV0g54PaPeByHq8s1kizC5dXfATNUXV7K0JVbwNnP7gdQSwMEFAAACAAAAAAhAAAAAAAAAAAAAAAAABoAAABwaG90b25sYWIvY2F0YWxvZy9zZW5zb3JzL1BLAwQUAAAICAAAACEAwmNPwYEBAAChAgAALwAAAHBob3RvbmxhYi9jYXRhbG9nL3NlbnNvcnMvZXhhbXBsZS1zZW5zb3ItYS55YW1sXU%2FLbtswELzzKwbwVVbph1yAPfWQ3FK06NEwjLW4tghTpEIu7fTvC0tO4panneHszOwMz64VFwN5ZA45Jhxjgo1tBgUL4Sy5xo8oICQmjyFFW1r5BvIeF%2FKFMyjn0rOtlbMG%2FEb94Hk%2B2c1JBerZ4Gmi8XtK%2Ba4s5za54RZuHlqs8PITJx8P5Oe5KyKc3ptdnXRIsQQ7D6U%2FcMJAiXoWTrlWZxesea8CzOC8L1kSibvwvWmFEAVCZw44ptiDwh9YEsodsyg1uDf2%2B8FJ2xmsao3Sq8Q5%2BjLV3C611hUWjdY73DKuzkpXoWN36gQuYHTIimy7PzjJBoulOhbv91f23mChtdZglZjsPkSX2WBZ35hXNtD1Bvxl6PDvm2FgOuPXU%2FVxHgmaRiP06pX3bUkXNv8tjXu5j1E69GR5XgaMwtHMhdODB3ClC3sOJ%2BkMtmt9Yyusx98KzR1P6gqbO97c8dcR7xRwO2Kr65Uez6ig63XzMTaf4%2BZT8MA%2BaFfTuFN%2FAVBLAwQUAAAICAAAACEAvpEdodgAAAAvAQAALwAAAHBob3RvbmxhYi9jYXRhbG9nL3NlbnNvcnMvZXhhbXBsZS1zZW5zb3ItYi55YW1sTY5NS8QwFEX3%2BRUX3NqYzjgwZFZ%2B7hTBpUh5Td7YwGsT8%2BHMzxctgtt74ZxzgcfgaogLCQovJWYcY4aProAWj8qlFo3nWEHITIKUo2%2BuHkAi%2BCJpXECltJm9VsFb8JnmJNytuG5UC81s8bDOeF0tt8pzcTmkH7n9V9HrLZ5e%2FmJOoU4Qyh%2BMFM4saxWhN90YKm7u77RSv8%2BQQnWTxU4btFllLlHaCn%2FrN3tzid5srt8VeTeMoRaL3qhjExlOLGKxNcYYsMpMflhiKGyxB6tPtjB6B75Kk%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%2BcgbRJAAAATAAAAB8AAABwaG90b25sYWIvbWVjaGFuaWNzL19faW5pdF9fLnB5FcbBCYAwDAXQu1OEnBTEAbqDQ5T4SwtNKk08uL34To%2BZT0jN1sQT6XgsfKcYHTOb4H%2FF1NwJpUDCabUR9CKo6d2hsMC1Hcy8fFBLAwQUAAAICAAAACEAaMgX0dMKAAAlGgAADwAAAHBob3RvbmxhYi9uYi5webVZXZLjthF%2B1ym6sC4POUvRMxun7HBLsdfOusopx96sxyknGkWCxCaFFUhwAXAkWaWUn3KA%2BDi5Qw7hk6QaAEX9zDp5sR5GIgE0%2BufrrxsYxtg3jRWq5hKWKBvUBgqloeJaVApqZXGu1Mqk8J1BsEusQGkQZa20f0wHg7slGgSuERqNBmvLSWAnLiEhsOBy0Uo3YBIwitZCrhbGalGXH7zGAnQrcUB7H82Foq0XbhHkCo0TxZtGbmGJGtNOTV7nUHHbSGWlmJMqA1E1SlvMQdRG5E7XXlhGj1unMn%2FgQvK5RFiTyAfUvdGg23oQSbXgUm7hQXAnZXYYn0GODdY51ostlFq1TQKidpPmWq0Narfo1VblIsc4Gcxb2%2B9Mtui2tqLCXo5AA6qAZqmsqiWfpwPG2CBYAxoHhVYVNNwuydDw%2BhW3y8Hg8y%2FvXtx9%2Bc3XMAKN6UJVjZAYaXY%2F%2FjQavxj%2BjQ9%2FuBn%2BbjK%2BX2fpcHId309YPBgMcizAkgMiUjnOBgAAjLFvl2oNb1teW2FJK26Ag6m4lH46rIVdwp%2F9hC18AH%2FhskX4AL6rhYWFkm1Vm3TgxL3QpfGC6UP7ZCCFsWTqbBZJPkeZdJtt49mMQHY5kBA0K27j2Sw5SKMPpmVK89kftjWvxAI0r0tkCeQ6AbbL0ttiz%2BLZLIU7gh0WvJU2CANhYDbbZemH5X42Cwq%2FRtvq%2BkjnFzCbebClrUidA2az51DxFYJwQQXJjQXcUA4Ygq4qgMMCyV8KcmEaybcgbNo52H2HCHY4NlApr4HGhdK5gRGMJ%2B4FZUbnjwSui8oS1pwvD1q25PoRFGz3NqXfJvvHqz0jZ7Kff%2FyJATyBXFRYk34SjTmOb4kWOOTcLPtAeR1S3hA6ox3ros2yThXmos4yiIrKjm8mIAog1VAadJ7%2FsNyzOPWujk6C9jateFkL21JqACPYsMyZsI%2BDCygIcOTyKGiUgEGJLpdHX6saE2h4KWpHGaMvuDTYIVuqUqpySswQbRLYJnCdwCYovw3f7isIqpTG8PNBihpNeFj2D32OfKXKoVQl0BjQHhR0u1ZdWpBfI6WBa823Jk58zqiOcCuuV6jd6i5TvlBSqrUJgLgyjmMxc5HRK7%2BHqP1DpXJMgIMU5dKCqhGUXaJeC4OP5h3Zn0EjanuuXyO5qDstnQ1LBMMrBIl1aZdpL%2BPEdRnwjTDef%2Ba5W%2BVgRyI2jpUDyToAYd7LCcsllljngLXVW4dwErFo9QP2UykiWe%2B02WwXVgUZ2%2F1s5tbixmruVxtQxOQHMza9OB%2FVLqC%2FJPiBoN0Jp8TA%2FLGQXXAFccxRNXqxQfMrMcXJ%2B27DtNk6mHADjbReR0qFipfRQ8DuUXqVaLm1OnpIgB0SkiXwEPdLCbZTimxkcWNp7DHOeTjmHFHAkpuDZDfC4sAL7FyJgu1I8h7GO0cAEyfAiXYraDCoQ8gfESnwpplWaHkUpy5JYDQCRsNeuku1RtrU2K2keliTkMjNmM75YkUFu87dRk6oVy0UB3ZkYSHKBPgGRl5cOyf3mqgQpRE%2F4Cj6KIEP09%2FGnrTowzep552IfL6JE%2Bf7bdwxjfvbTyd41bwixiHEGUrwiFBPxL3bx6mwWJnoSKN3b%2BIlHHYisf1G9TQkxwhu4Snl9vE2J%2FqIBKIjnWLSCeu2Qs0tRpFPojP9LhXkGzfRaejlJNQZKD0q2Oc7AU8PKu1Z4pOKgjViwyF73IQzXzlX%2BUz%2BX87im%2BU7dGGl5tvT%2FbPwuBa5XY5uH9eFb1KDduopMepzpOPITXwxd3s%2BtytB29O5pRZ5dKdbTGC9FIvliM2VXbIEuGyWfHST%2FqafLgqvHTmgC2cfn0BzF9BxTBcdOVaUqaVCMpV8q1obhkJ%2B8k2opxoL1Fgv0EwLpSzqSGNhEljh9qh3fAHs9WEi861ex%2B6leMCa5Jh0LuZu4eNdIhYmI1Je4TaDHVsIS31GmqYJsKKV0j%2FsiaC5Aal4jjnMt26Tb1%2FeffcK5lItVj3z01bZ6c6%2Bq%2FWtqHYldKGqig8NNpyQnoM%2FoDzabjKsHvjts48%2Fzp7dPLtN4A2v0YjFKnt2c%2FMRe3cveegkq5w63dY2rf01W0lRgDCiNpbXC4zI7ITMOkoR54oRjFcpmdtEsYvWirLLxcc0UtiIJSwmYYdZvjF1CUerCzaEHbl3vJqMryhEVxNfCZwkGvEkSxPHn%2B5W%2B08mENExKIxSWOK%2BNhwS%2FqDI5KwvrPKI3dcsfaNEHY3ZkydwBLsEGEvg2qk3iakhfALD4bAHwPD3oJHn7jBDtcFARLtdQDymZYPB9OXXd6%2F%2Fen66%2BjS6Xz%2BN7831%2Fe7eXEfjvyf3ZuJeJCweTL96cffyexjBTrP37jXlF9dard9jGbCf%2F%2FkvlgANSF7Nc%2B5e%2Fuff7t39%2B%2FTwPhkxHLrJP%2F7E9iEJp6T21JGYK8gZhTMBkYGobUx22baROPZvazs5JOZr5LmD%2BWdifoffQyFQ5oFKjeXairoEbp07xmLyvPO1Z8wEatxQtHLcxOkBd0U33VXgHetxlWNjlwQ2rqlHuElAnDD5Gxd5OqtFInH1iCSdFxInBZ6OYEfC4TYBtmcZDG%2F3aYm%2BJRm%2FmSRw07NZ0MuvHI3g5lTiEYrcaq%2FgU7iFDEjSG%2FrdMxEXBv3x9qXWSkesredcUjblMNd84Ut2hysWP%2BKWK3bVq%2FAGRm4sdZ6MrthVAlQMb3sDjrUTj2pWeSRW3C6WkWbj%2B%2FVw8pQlYUk2OT1Dpe56IrqJ%2FU5V6tnf46nh2uB0LuY9mhyKcrEg3%2Bjk6JexetLj6U%2BiFhWXHZycIJ1Bx9l22xw420EtA83XobHeHxBE7T8dRUaw2x%2FOuhX51OdcWog6F1RsHDyOOzOUdBoUtLTbrbP1Nk6lWqOO4n1yMLhbuV4KiUAV9hQZxcXtibnuMnx0b65ZHDzuG2FxATmis%2BISbXONfHXyNmQUqX6ezwkUXt1T6d7acXFh3wRGXtypMo%2BYknxyYcCZY0Isxp0XnznpfutjRIV5HSO5E6RniR4%2Fxurs6O7C4sbD00XWMeNlt%2Ba5KNiTamwkX2B0tPgE1wwC%2BZ9OZzvH%2FnH%2FYh9e%2BErmqoHT2yyVtlPe2qXSkbcyO0P7qSmiAOYW%2BTWMbAnrzpM3OCWE7WRVyE7%2F5IonPym9vBfrOI6FdcdGMHfAZl0VLg7S6HX4Pb6ZpI7cDLWc5Jc4o1sgpUteix%2FchUni%2B5nd7ltVIXym8u1%2B%2Fw5Teqlef9enjGDcDR81CjSnNyUsPKnftDjcGBHz02NMVHnrm4SIwftddP3Y%2BcRnF1n2C5%2FQeOzCrnug6y6ZHq5AD1V%2F6lqBaC7mU7pmjU9q56FXgAb1SSvpWtWuSWV%2FDK1gAq4X7DtWlqYp6%2B8ZT1vebs%2FMXfBSr9ft8O5W8v9ojxNqBlvju%2BPZ7KK%2FIW2Cje5btfaUildIV68Ux75S0J1z76TUMZg7YGO9ULmoyxFrbTH8mMWPHMm2yDWxisc2PZGL6jSneHSTPGaIHkMePZIJh%2Fz2uZAAew7siDbdnX6%2FU9POpTBL1Cymfj%2B8pdZY2JZywb8%2Fup%2FAXLj%2FQ7gLjmJ8FZ6vCEB5Cq67ZeGl54KLKw4Khl%2FvVd9DtCOT93EK17sT466ssBKvEri6iuP9dQq7IHrfzXMGxfu0F69aO17hllj6gIKC7c6YjWqg3%2FUIjPR15MKUhlIW%2B8CHLFWtHfwXUEsDBBQAAAgAAAAAIQAAAAAAAAAAAAAAAAAQAAAAcGhvdG9ubGFiL25vaXNlL1BLAwQUAAAICAAAACEAyOkX2mYAAACBAAAAGwAAAHBob3RvbmxhYi9ub2lzZS9fX2luaXRfXy5weVXLwQ3CQAwF0bur%2BPI52g5ogQYQsgxxYMWuHdkR9aMcOc8bZr5GL8OM1UZhi0Sf%2BjKUeUVWY2aiLWNif8cRPvTR%2FFza6G568j3ywKr5ka9mV3%2FagvIkEtExRHDBjf86L%2BDy5Dv9AFBLAwQUAAAICAAAACEATZ8LMzQDAAAPBwAAGQAAAHBob3RvbmxhYi9ub2lzZS9saW5lYXIucHmtlM9u2zgQxu96ig%2FqxVYdORaK3UJGgDXqHILsuts07aGGYdDS2OJGHCok1dT7IvsIvecR0hdbUJIdB3GLHqqLIHE4f3%2FzhWE409IS9BquIJz%2F9XGCUfL6NUrJJAwyocgIKJ1TGeP8S1UKyZRDMkq5MsJsh5bYajO0bGKVx2EYBoFUlTYOXKtqC2HBVRCsjVaoCu00l2IV1yydRWdYG9oEwQtcF4TJ9A2Mrjm3cBp3hS4J09kYNcu1Nqo9k7wBGaMNCmHxWRgpOCOMhqME09nDfRy8%2BzCZXV98mlxfvJ0tP06uLiazN%2Bc4Q2%2BEIUZJH1ETNZ7OoigJguCP5isrKLvphXMqKXNG8yIcIJznciOdKJdcqxWZBYY4MOgHOa2RC3Oz3OXRs3KjxNKQyAe47KcBAIRheE2%2BXFE21uCm8bs7Kbx597Mqa4vbWrCT%2FwonNbf%2FB77t5KtrPE7Mxra%2B%2FfMY84mrOS3ivc1lCru1jhQ2QjLm09nQHzfnV%2BRqwwcOp8%2BT7ILkD%2FeY08P97urE2lrRwdWrx%2FiC8yOVQBiC5Jwq4pzYxXj3zGbvTdrd8Ae4K2RWoNBlbnFXEDfUti5tJdhCaUNwhWCIla4dNHt8ukQP%2BU2PA9z1Yp2C1GfhVyFNTpMRHr4m8Suc38YYvdrNs3mbpm84mHkUJXiJ4wAOcRlFST92uhdSFCVh%2Fxl77Y605O0x88wdPfhZQC2bnqqX1QDkxADfIfS93LAoT5w%2BaXtq%2FDy8OPg2VvILlVhrAwFFgrttRqZrdkeZ9AHTQ1uLikznaF4VizFsJkph4J0aI7aPrJITaUtOrUDrtcwkcbbFnIZVccD0L%2Bf%2B%2FewK81wqYis1l2TtYgArFMEWoiKvZ76u77H%2FZyucLVRYUanvYIWrm1byGH9raa1m2EK7NtcxWLeSkNXGELu9L788rD3ZTooSrPmk2wTptui5nZ74lA%2BH1P81vP%2FW8J6MxvhHMFmZ3aTJ6envT%2FBX9ZJw5qeFqOkL8MLPhEXpZ9DY%2BFL34oiz9k6nv%2BTtu56kj0LTyIs39ezMaYFv%2F2GEVhN2G9e4GYKr2N4a1%2FuRBOPl0yT67QY%2BmXLYD%2F4HUEsDBBQAAAgAAAAAIQAAAAAAAAAAAAAAAAARAAAAcGhvdG9ubGFiL29wdGljcy9QSwMEFAAACAgAAAAhAHA4EKSVAAAAMwEAABwAAABwaG90b25sYWIvb3B0aWNzL19faW5pdF9fLnB5dY1BCgIxDEX3OUXISmGYAwiewQOIhNppnULblDaCx5dpnZW6%2FS8vj4guRYNtJ1xc0RXFoxf7bGjygs3lJhU1RJ2JCMBXSVhWUcnR3Gfp6jyEkIpUxQMgIi6uj7wJvPnTmIP31VgNkrn3WDz3y8EfTpLTGuxPmszr8%2BsIwGxiZMYzXjukryQNi%2F5H94t%2F2Z3vYZrgBm9QSwMEFAAACAgAAAAhALPHYO6rBAAA1A4AABkAAABwaG90b25sYWIvb3B0aWNzL2ZvY3VzLnB53ZfLbttGFIb3eoof7EZUKdpWU7Sg4KJeJEWBwos0O8MQRuQhOfVcmJlhZAd5jL5IsyqQXR7Az1TMkJRJ2c6laIqiXMgyyTmX%2F5zvzCiKop8lq2hpeUEoqHE1dIlS560FUwUsKasNHBcuxdPrRjCuqABXEHxrmLk50o3juT0Kaze63IS1qSzSKIpmMy4bbRxUK5sbMAvVzGal0RJNrZ1Wgm3TVnFn0b%2FYGqpms1lBJTY7ba64qjblRrVyS2Y%2BfEkgWaV4yXPmuFZxNgOAr%2FD0ZYr5SZzhfLPDKc4xP8HXeCPfxAmatuFiug4naVhoyLVGYbCORbeObe186ieezWY%2F%2BgjTvKb8ah5dXEYJhk9BqnL1ZRSH6CvSkpzh%2BWaqzGNJJNiK1mwKziQ5Mn1OURS90I4J8EerVGoDpsCE0DsqghXYRjvM9yGgq1GczoLRM1PZzry%2FhngylMtegHNv3dUEQcomgwe9%2FY1yB%2BbAVckVdze4KLgkZblWgqy97OT01yS1DII5MuxQfrnG8d8xPhHKGzcVWQeW59Q4thU0EmF4DTmGAiWgtEqxQsOvSdhelOehCUa6dLpPxU5QGq0cnMaW5Ve4aOXlIKq1raTR%2Bp8O1E%2BgNApeloblIX2Pl9JgWzImCGLXD3fpem%2FTSiYEcq0ITFWCbIoXdZ9uZfTOQnBFzIgb7LirUVBHYxfiGN%2Fso%2Fz2qpQZrOSuzlbHx9%2BvUVNeu2x1fPLd0J9j9lZxBjfS7RSrwGI%2B5my%2BwuIz4MbigIzU6XnUyuhzaBzp%2Fsk87tgr6qx8Ooy6BBs7WwouuaMikIT5y5YZR2bpTSM33JHxKf5PuLwTLBt938fMq9rtIfxS1J3taVqWhjql1n1xyBifVD8o28YbZOhrEiL2Lv3f4G9vs1s3f85uBPGqjtfjBjg%2FwylOcOTb%2BnyzixPwlNL7pCZ%2Bz2Tc%2FGssfhNneP%2FH7bs%2BtLP3b%2BNuKBxGDCx%2FQI%2FtE9y%2B8zffv50Q%2BwSLcUU%2FD98FVh9l9oDWbmxt%2FEFh448ec%2F%2BRoODWMZVT94Bdc3tH5rNQYlvzMnQrQ6O5ch2QfjUNZ5kEhgRz%2FBX5%2BvvO9I%2FhrT3IoX%2Ba9e%2F4SmJLbkekhqORH%2BOhH9AIpggX4a1hl5GGFXeE3E8g29%2FzTS6n8UCrcKP3NKh0tx%2F4q7WEmokOsoKzSismAr7%2Bxk4b65Bro8g8xtzZNQ8rJgK6mrleww%2Fx9kwwN1ai9P%2BP5ZgrjZKTKJC35hVzraF4PUrR1Ua3VX0nd0292xABt56brd%2Ff%2B1mgtPFs3YyVyUk5%2BufQeginJ3GG299f4xQGlivc%2Fjkh5H5lsYBqUstVaN4AgGFFFMcfgGHA4D4Skl13JEwT%2BDATv%2FQno9EhvlP1iqixe0b8VOBdp00H74M8TAOY7vePTuz91P%2BPk9DvDcHvhWf38cYPuHALSz0vo3ATWA1ieY2wR0h2cxDcgcyD0V%2BZpIDQskOGea9NOBv6H0%2F35uIX7fhv42zf6xvJrnHax32EVdf7fsLjFAc92e0si4c6s2v9yfkhitOwYbi2oAlRqkmZyT0%2FqkklV1y2ch5cJjhJj2N%2FMAzwGFZ0dn29onj2F1BLAwQUAAAICAAAACEAAAAAAAIAAAAAAAAAEgAAAHBob3RvbmxhYi9weS50eXBlZAMAUEsDBBQAAAgAAAAAIQAAAAAAAAAAAAAAAAAVAAAAcGhvdG9ubGFiL3JhZGlvbWV0cnkvUEsDBBQAAAgIAAAAIQAQ9dbURAAAAEsAAAAgAAAAcGhvdG9ubGFiL3JhZGlvbWV0cnkvX19pbml0X18ucHlTUlIKSkzJzM9NLSmqtFIoSkzJTMxLTtVRyCxCsAsy8kvy8xTSckorFDTy8ksUKlNLFDJzC3JSc1PzSlJTNPWUlJS4AFBLAwQUAAAIAAAAACEAAAAAAAAAAAAAAAAAEQAAAHBob3RvbmxhYi9zZW5zb3IvUEsDBBQAAAgIAAAAIQCyf%2FC%2BuAAAAF4BAAAcAAAAcGhvdG9ubGFiL3NlbnNvci9fX2luaXRfXy5weX2PQWrDQAxF93MKoVUDjm%2FQU3QZgpjYGlugGRnN2MG3L3XwItB2%2B3nvfz4ifnGp5lfljRWSTKtzBUuQ2aVBLCNMbJmb7z0ihpDcMiyzNSsaH3099P5kQPJi3iAJ60iWaBN%2B%2FuEs7Mk8xzLwqX0EAID4qKZrY%2FoBpckmbac2O9fZdOwOZtxLzDKQxzLxK5qiFErmlFZVqkNU7sIlBKKoSgSfcDs4%2FL8fX234tnCGb7%2FO8Jdh7MI9fANQSwMEFAAACAgAAAAhAGPQg%2B6MAQAA6wIAABwAAABwaG90b25sYWIvc2Vuc29yL2dlb21ldHJ5LnB5ZZJBb9wgEIXv%2FIonTna1daOcKvfSSlGkXnrINYpW1Ax4VAwuAxttf31lO946KpdBw7yZ7wFa6%2B%2BT8Rw9PKWJSr7CUuYLWbicJghFSRmWJ4rCKUqntVaKpznlglin%2BQojiLNSa%2F08ppJiMD%2B7GrkI3gprJq%2BU%2BrrEbhhp%2BNXo50DRl%2FFFn%2FBv3ypLDo4p2HNy5wvTa7MxnIX%2F0AkuDSact%2FK2VwCgtX6sIcBEX4PJmxrJYVHDDDmJIEX6z0ynVv237GXrtKzDtH5XvLIt4wkjsR8L1hbGp2gCdvDupj8C9iDnaCh8oS2PLX%2BQrbonKjXHA8Tq572P52zsyxdUIXQlNdqS1y3cAkM%2BE8nuRqROdOj1wFJSLpziR5eJ0OSFKHAkk9sFSBa2KmRhCjg6jlyuJ0hCGQk8GU%2B3ZpalmDgQ6Hc1QdaKo7MOP1LBxQS2K5tjGelK65gb4hO5HjJxGfv7u7vP%2ByOuMa9XgeYeHxDnzuShmNgc%2FwA%2BbafvfkK73cn%2BsoFEdPuWzcbqVv0FUEsDBBQAAAgIAAAAIQDA9mYoCgQAADkKAAAfAAAAcGhvdG9ubGFiL3NlbnNvci9wZXJmb3JtYW5jZS5wecWVwXLbNhCG73qKf5iL5DCUxGlTjzKeqV3n5I6b2L1pXA5ErkiMQYAGQNvMtZee8xK5%2B%2BJLT3qA9B36JB2AEiXaSg5tZqqDRArAYvHv%2Fy2CILgkabjlt9w2sIUmUyiRhcgayUqeQjOZE5jMkDMuI7y9rwTjkjJwCcEXmulmbEgapcdG6qjMoiAIBgNeVkpbyLqsGjADWQ0GS61KVIWySgq2iKTihiLBJTGN9fyM6evklmnOZEpPF9SSW7OZWWvKB4PBj%2B43SgtKr4fBnASlVit5hTHm7cKrIMTOgH%2FLeM4tE4msywVpP3k7YTTIaAm2MErUlhKz1Sfp9BmSZSEMz0uWaGJZiLPRbAAAQRC88%2FsaVKRR8XsSYBZ3BU8L2IJweX4BuqmZMJhGA7%2FoWOemXe4%2BZNkMNzWTti5ByyVPOcm0wZzGVXEVdfO228%2FgvuEFxZx25pzNYBpjqfTlw%2Fz0fOyG%2FfgF2VrLnY2P12eG2ecJlHVShaWL4tNoMzemLmknxiUrCaXK3KnNs2I7h7QLd30022%2BkdZLLGai8ZdP48HAWT%2BIpVp%2Fi6Ae8vYkQv95o7n9feG2PMEXOb8lg%2BPlx9fj5z2q0esDff3zE%2BtU%2F%2F%2FV7tnrAESZvYNk1%2BcpUyp%2BaoJWyrYYv8GtBmI5jj8B0%2FB1SpnWD1ootGLR6wIJSVhvCO8WNURIbC%2BMIJTHZBivrhBIn4BEm0fc48B6OCC8hq8jcaDucRPH2%2F4ODGC%2F7RAz7lhv5sNrXEcMu%2FNhZaBRZNQyqIhg9o2QvGP%2BZmHXHSHzHcMlUiWE2xNdQOe11GS6RnYTw1Btma80sVxKZupOwqoPHFbhz5V6CNpvP1gZ8wuJO7NbK%2Fzt5fR3m2clV6PCJJ6tHofLpBGrZOtQfBz75LxH4s7ojDZIZuNmv2SbauvV6Wt%2BgrqpuXRfMTduRK2UVS7ltIvwiReOtzWW%2BrAXuCpKd7m5ntlC3LVZPa%2FVt0D8Msfr0Oup3AJ%2BmY26TyPhfNPJdqDw3750juG2G8QQHjlVfk2FbBUdZxksXXElBxgSjqGS55LbOaBQiyE72INina4OQc0qyVDpZ1kIkJmWChv7xjoQIwbI0WXBrtvxc7jjMFsyiZFVbdLfslVvW1QxKriEqeF6QsTg%2B%2FQmpymgvQ92%2Bs32xelbf5DXzETV5wZ1dJOY9aa5CUJRHmMabJtcb%2FhIdu6c865P0zP3HyMjwXCItFE8phFQWzN9oSqPSqiJtmxk%2BkFZQy6Uh65u4p8s1iC3kre3JgO5ZakXjWoeTC%2FFv0t8h0wgXxJwqJWlmUJuaCdEgFbwCMS04favrrufykt0nPpEjDGMcrKXsXNrZ5OvexCtMR5s6nJ73r5LNDuOtD9pwTvxgNPgHUEsDBBQAAAgIAAAAIQDzA21qJQIAAP8DAAASAAAAcGhvdG9ubGFiL3VuaXRzLnB5bVLBbtQwEL37K57CgV2pTSmgIlbaA%2BoeEIeViqCXqoq98Wxi6oxTe9KSv0dOUkRbbrbnvec3b6Yoih8tITluPCG1JpJF71jwk518p8YliSOOIaJvgwT25lAqdRkGFscNBnaSYCIhUW%2BiEcLBJIJ1HXFygRNSgLQ0ojbMQXAgdO43WQz9RqlTaE2eaomBtcZKa9J6vZk%2FO22IKWta1K2JDaE2MTqK6QR1dkC2nBRmazO%2FbycBx7WzxLLYfsmwrnFifMVDd6A4M3f7iflld4kwSD8I6mAplUrtKLoHsrgfDIsTRwnH4H14BBsZovF%2B3MzFoQMdj652xPUIl3J7Z9kSDFuVxiTUoTGO59puf0Zal0rtg9BmCv5tgqWjGbwgPsU%2FJJqUtJ4mIS2BPHXEYuK4ZFPiK0WaUaojwwmBJ9wULlZmTmB9skzktUT2FIlNRxYSoPV9lcUM2yx7rfUMsHR0TBbWRarFj3CMX2HwtMzaCJzgjqhPeAzxznFTqqIolHJdH6JMbSo1RGqwnS7lv8u2Ws%2B1cv5nVfxtYYubp%2FMttqBi%2FRw4jzrD5lMG9e1L1PPRY4ub5y%2BZtdsXa6Xe4DslCZGWNX9sQ17tZTqTnpNpxyMdKebMcq5p7A7BQ5MuXzWyBF4tgW9xXl68e3%2F%2B6eLiw0c6Pf%2BMS2xxX71q7anv6iF4%2BQ%2FtW87jOpu%2BqrDFxL2at3VUqqqM91Uu3BS5VJyguKqKW%2FUHUEsDBBQAAAgAAAAAIQAAAAAAAAAAAAAAAAAaAAAAcGhvdG9ubGFiLTAuMS4wLmRpc3QtaW5mby9QSwMEFAAACAgAAAAhAOl6sF5RAAAAUAAAAB8AAABwaG90b25sYWItMC4xLjAuZGlzdC1pbmZvL1dIRUVMBcHBCoAgDADQu1%2BxH5go3vYD0S0i6mwwSpAtpgb9fe8dN3PFna0VFYLog5tY2HJXIxgvBB%2BTD25V7Tg3XIZxLSdBt8FuyxfB8yUUFcYsn%2FsBUEsDBBQAAAgIAAAAIQCozWIDDQgAAHMRAAAiAAAAcGhvdG9ubGFiLTAuMS4wLmRpc3QtaW5mby9NRVRBREFUQZVY747buBH%2FrqcYYL%2FYikzv5g7X1pcset0s0AWSXBC7CYoggGhpbLFLkSo5stfoFuinPkDRJ7wnKWYo2U4uQXoG1pZNcv7%2B5jfDfYWka0169g5DNN4t4Kn6PnutW1xA13jyzup1dly8VFfqMlv2bavDYQE%2Fd%2BjAtHpr3HYWK4OuQiDv7b2hBVTaVr3VZLyLBeBDZ7VL30C7GgJuMPCJqLKfemp8WMCScKMdvECwGhp9wGFlhq029gvr8CzKT6pGZbXin%2F645b2q8u119tJU6CLObh%2B6gDH58Opulb3Fv%2FcmYJy9MJEW4Pq2O1w%2Fv1JPf%2Fh8qTOOrp9fqqff%2F2rlcNCtvX7%2Bg7o8Lb05UMNKrp9%2Fp66ushcYq2A6dnp24x2ho9nq0OECCB9o3upwX%2Fu9y7KLs2hnP301qDCJ6KIPsQDfkaliAUHXxrdI4VCA8ybiVMGtrhog35kKTIQ89w7h1aALOr3FPAfjoLRmHXQ4zEugRhPv1eA8YQEarNmhfFl7fw%2BTVgfT%2BqlkTkNsdEC9tgh7XItE0ATUIETdIpBpUcGqwU8wkESy3hiq%2BdHfeVnA3lADvTMUYcIRn36GEJh8CLiJam3WHyfj03Sqsqwsy2x0A%2BTFVrBFESaqrafwy7%2F%2BC75DV0BAXReAtaECQu8ywrazmjDKST6j2lr2V747ADUmAnmIpANxZHAvmzI2P%2Bk66hudgU5X9xyOyRNIqZpbdBECVj7Ukb3%2F1PlKk7Z%2BO59mhJGSJfzqd2widAf%2BOZN4q%2B6Q1iTev%2Fz7P8B5%2FfPq1UvYGMsYYZV15nuay4H5aN5xT4SD76HVNcLEeWJztoamEsXs4gKWSH0HE%2B8qnGbZnYukrYUP%2Fe7jpCHq4mI%2Br30VlY4UtFWxmfe7%2BbRgJY6FsbKAnY%2BGfDjAxtsaw0Ky1Pk9htigtVm%2Fg3hwVdK64ihXATkPUKodul2Z8IA7DAdqjNvC5BiwAhIS%2BZM668madTGEaargr74HxwfBIdYcEl2R2WlCMPQjlCmsJdQeo4Beid%2FvlnDja8wyYbQveqES22mJfgGotupUQEOqowuqrUuY3CN2YFJFiAEIpNcL%2BU463kdJrk%2BqGJySnTGQeb7C0BqnrWT5be9gpeN9nsPEB7ihYJ8sG7OhJ29kPefFuDjtmy6y7FEe4RHeS2VT8vgxe5zNZvKXPUJ%2BW4uNJiZSgEfxMR5LiFOaws1PB98HWAe%2FjxgWUPka4QlEa2oMUcFS7zDCWlf3HHWWUHIsFOuE%2FJ3B%2FaealrrFAtY9geZ%2BALrrFuC8yC3AO3sQjixGDcIInPA4iFwyxE8yGbVnSJ%2ByjvfBCKpOFfGMt16rhlo7mvamX1sTG2CkJ94gD0Hva%2ByoOZMicEyqyEOp1HzcNO9YxFlVz0TOvIQJPnQYTIuOHUGENVq%2Fnw6aOWFS9axlgOYA5RIes0zyk4vjEsA8QeU8O%2Bc5YQouuUwX87n1lbaNj7R4%2Brur35cKluQ7Ptmm2hIgBh2bWaUdmMo78JsjPMdSphGGnXZoT%2Bi7mTK7a3cvsSpns72mqikLyZhQDJMsSxkKC7SNHvqulio%2Fgn40XKiXqQBHuI3n8IHQcd%2BWA0wQ3aE8tqUoOOG%2BJ6UWB8ylNBbAKiOeik7Be0ON72nUckTp2PpCq%2B2pUzKOFskVcYwTsfGh7a2OUJuAFdnDmZeJS35eR1Mb7bLsSkGeC2skCmGk73RvKc%2BlcKvG%2B8G%2BX9NN9pRPL5HIuG2U%2FTe%2BbblJHqCz%2Fda4OMgxA0%2BzfeikLef54OKfgqm536vsOwV3DrjFxkFoAdQHoaHBzE88l2aVpOS5gpeomcUEe9REwLajw4KZZWNcfaJuQxHtRsFKRG82kERz8%2Fpcqsqyt6h5nkgTxFkOOMMKVl5KgofJPA9m29CssoYZZiwBdj9Noee23iX4DiFg8BZQCnJVKSjm1XVP5B2sPZFvQaRPIe4NVQ0TGdIe0aXstwJhnny6DnYG9wxWD9iuka0PvXPcpcai1M5Twy3IE6bmV%2FJbMidLqPpi58hY0UJmkwbZoAX84fJS%2BmR5bNFCezoK1X2hsw4c0niXpjQO%2FBeV8XBwAbNrOLEjLzA1ps68RFenwZCZVUaMiczhBaxQt1FIgKFbmc6g4xbTry2mFEXJ2jFPoXeSeWrQHAnrxO3GVbavsVbw2kOaocGHEdgyPkpDx3poDUxajjA4JKi8c1jxnmSQWFp5R9qMvew4nnF3UVmW57dHYl4cyV7luTAElDNfcpXI%2FDAUZAHlGM8SnG65I%2FCYLdr0hjChKvUHQ3YADPPefmhDxtX4kHqPLJbCU%2Bpv0bsSJnKoEHkFxHTFgm5gxzHWg6UQDXHkNRfeuZQNR40ilKdmlOjwxzSd08EyUnkSl3wcXf9tSIKZh283wONY%2BXqcnLMbHqzL0%2BA9DN2l9JERps%2F4cnE9fyY3mGtelrDYsS1tgnfUauKYT0rDy6VEjx%2F4dsCfm94JJsp0l5AcqNRUtjwK8T2kLEt%2BG5z8h0oV%2Bs%2B0sLa%2B4p5xIr5yebv6yxu5INRemg6zw7BzvAANfKKtd0ND6HdQ%2BbbVro5fmYTh%2F3pdHIl%2BPrRSUYhuZ4J3DOUxawPxCXd9tfbhIm0Y2Hfy6UA3%2FUwWP35dFFwILZ4zuUx0v42QLj670Bh3xk1nojDSt2MlgxUf0nUNz4Zr2fU3DvFeDTV26Gp01eGI4OEfCVn26m6VJrkPL%2B9ubl8vbz9OhoepghtDWMPeh%2FsIcgVIdMcd1SYBUWX%2FA1BLAwQUAAAICAAAACEAH%2BMT0C8EAACKBwAAIAAAAHBob3RvbmxhYi0wLjEuMC5kaXN0LWluZm8vUkVDT1JEjdTHkqNWFAbg%2FTwL9JDDwgsQIJCEAlGwoRBcggTcK4IIT%2B9yjdum3ZoqL%2B%2Fm%2B0%2FggArYw6aKbz%2BjqGzKPoo%2B0Ix1RUyx3B8RXCDHZPXeO06eV1PVkJz1gJ4Sp4xV4VLqSMtxufa3pYTRpPgD%2FYMlcR9XMH%2BHyhcnHSRjzk38Gun704EXmsYd6FTAKY45bXXt6OhSaz7HC0ZzlPBGrUDTge4nmOIaVQD%2F64nHH3NcV58hPKhuwtFoZvHAmwSQIKtrDQjT6iGUvptnpLYvJJqJhMrFSJ78nxm3LxnqyC%2FmwSfn3nvun9SIjq9lyUsR71wAnCESfMrro1w7izlGssSbjA40HWz%2FDfn1%2Fk8rbFGfut2iMyO1AC72%2BAvZmGh3OQaiJ0yihZLhlHF9gwQB43j6%2F8d87aZ5Rqw3idcnrhBhVCljwOheb0PTRDQiu5fiqVfbvuJXMcBo4ksMrGD7btP%2BvQTKWVBO4cYyobWtoAM2lOacj22rQXavvO4J%2B1h88Moxbv31gAokfQubMuneuWz2VGjV6idbr0fyYQs4tOmpu4gynVWBfySsxDkx6tjlLvZlIDVIivh3qiC%2FoOw6Iv3kq03j99Qhr2OTUEXN5dSRjWs4FqW3LWKcwHhupTa3lVJ7U0CwdcK19S5kZ%2BZuOrQTenk0cNw4RQGrl4nZtIekghjHievqGlh24F1lYjIaYtmLTPqyuDwh9ap0YREEM71r00SpykQBo1OKActgJCV%2BI6uyAXG7LrKHu1AeyMKwqmuGiDsfXRTWPbOpDolGnrfuWS%2Bfln99jBgpEPxKhKj%2Fzfwa6bBlqzTy4jy5jl1op4mx4c%2FmSW2RyDl9kZn0cX%2FQ5vqB0e%2FMDCZDtwItIa4L75A8jsQyUFvpmT%2BKELYtMnLPOLY6002xFEj49axiNC%2BuV4Lmj35GIP2kGF5RL%2Bgu6LId45Fj1D7O7jZguADrUbNHE%2B38cEsX9qC52PpQ2zgtYQ36dn7XcUx61rQM%2BsuJbWg7e2HbtieuNLYooipmaZAoufbZSqjZxHh2xf46wHfkM9iSVx7BjbmfIz67E5bRZyLJbrNcEhhRGsXx6deZfnUfGP3ln%2FK3mYNf9a5MxUVTZov3rY1vvDC0iQ7Eu9A%2FosXbCS2xSVVFY40N75xHjGf47yYCbQbbOm4SsGK9q9yd2VSszFBT6iXV27AIhihC2dxvqIGMBY0Km1PmxwxGceTaHZqyX2%2B6Jc8vb8JPNbsX2Vh2N9PN3edQIQWkEcutD5Kx1sCeZUMCIwlqdTQ48UF%2BEB9p2fV42WTwp6%2Br6uHTnZyNrBmhYcyWPc1wPL1Y%2F5lHd83l9l1qZNkwVLJMiOdCwITVML%2BppupIiuRInzC1KIhqlluU5sErCLuTjD8Uco4Oo3wIdnhMIhC%2B%2Bm7x9h3GMNyq92%2B0pW5OloJhP%2F4EUEsBAj8DFAAACAAAAAAhAAAAAAAAAAAAAAAAAAoAAAAAAAAAAAAAAO1BAAAAAHBob3RvbmxhYi9QSwECPwMUAAAICAAAACEAG3hheMUAAAA%2FAQAAFQAAAAAAAAAAAAAApIEoAAAAcGhvdG9ubGFiL19faW5pdF9fLnB5UEsBAj8DFAAACAAAAAAhAAAAAAAAAAAAAAAAABIAAAAAAAAAAAAAAO1BIAEAAHBob3RvbmxhYi9jYXRhbG9nL1BLAQI%2FAxQAAAgIAAAAIQDLVDMCrQUAACwOAAAdAAAAAAAAAAAAAACkgVABAABwaG90b25sYWIvY2F0YWxvZy9fX2luaXRfXy5weVBLAQI%2FAxQAAAgAAAAAIQAAAAAAAAAAAAAAAAAZAAAAAAAAAAAAAADtQTgHAABwaG90b25sYWIvY2F0YWxvZy9sZW5zZXMvUEsBAj8DFAAACAgAAAAhABBCMBeJAAAAqwAAACwAAAAAAAAAAAAAAKSBbwcAAHBob3RvbmxhYi9jYXRhbG9nL2xlbnNlcy9leGFtcGxlLWxlbnMtYS55YW1sUEsBAj8DFAAACAgAAAAhAAvPgQZ8AAAAlgAAACwAAAAAAAAAAAAAAKSBQggAAHBob3RvbmxhYi9jYXRhbG9nL2xlbnNlcy9leGFtcGxlLWxlbnMtYi55YW1sUEsBAj8DFAAACAAAAAAhAAAAAAAAAAAAAAAAABoAAAAAAAAAAAAAAO1BCAkAAHBob3RvbmxhYi9jYXRhbG9nL3NlbnNvcnMvUEsBAj8DFAAACAgAAAAhAMJjT8GBAQAAoQIAAC8AAAAAAAAAAAAAAKSBQAkAAHBob3RvbmxhYi9jYXRhbG9nL3NlbnNvcnMvZXhhbXBsZS1zZW5zb3ItYS55YW1sUEsBAj8DFAAACAgAAAAhAL6RHaHYAAAALwEAAC8AAAAAAAAAAAAAAKSBDgsAAHBob3RvbmxhYi9jYXRhbG9nL3NlbnNvcnMvZXhhbXBsZS1zZW5zb3ItYi55YW1sUEsBAj8DFAAACAAAAAAhAAAAAAAAAAAAAAAAABAAAAAAAAAAAAAAAO1BMwwAAHBob3RvbmxhYi9jb2xvci9QSwECPwMUAAAICAAAACEAFoPrBUQAAABFAAAAGwAAAAAAAAAAAAAApIFhDAAAcGhvdG9ubGFiL2NvbG9yL19faW5pdF9fLnB5UEsBAj8DFAAACAAAAAAhAAAAAAAAAAAAAAAAABYAAAAAAAAAAAAAAO1B3gwAAHBob3RvbmxhYi9lbGVjdHJvbmljcy9QSwECPwMUAAAICAAAACEAmh58BkoAAABJAAAAIQAAAAAAAAAAAAAApIESDQAAcGhvdG9ubGFiL2VsZWN0cm9uaWNzL19faW5pdF9fLnB5UEsBAj8DFAAACAAAAAAhAAAAAAAAAAAAAAAAABQAAAAAAAAAAAAAAO1Bmw0AAHBob3RvbmxhYi9tZWNoYW5pY3MvUEsBAj8DFAAACAgAAAAhAC%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%2FAxQAAAgIAAAAIQCzx2DuqwQAANQOAAAZAAAAAAAAAAAAAACkgYkeAABwaG90b25sYWIvb3B0aWNzL2ZvY3VzLnB5UEsBAj8DFAAACAgAAAAhAAAAAAACAAAAAAAAABIAAAAAAAAAAAAAAKSBayMAAHBob3RvbmxhYi9weS50eXBlZFBLAQI%2FAxQAAAgAAAAAIQAAAAAAAAAAAAAAAAAVAAAAAAAAAAAAAADtQZ0jAABwaG90b25sYWIvcmFkaW9tZXRyeS9QSwECPwMUAAAICAAAACEAEPXW1EQAAABLAAAAIAAAAAAAAAAAAAAApIHQIwAAcGhvdG9ubGFiL3JhZGlvbWV0cnkvX19pbml0X18ucHlQSwECPwMUAAAIAAAAACEAAAAAAAAAAAAAAAAAEQAAAAAAAAAAAAAA7UFSJAAAcGhvdG9ubGFiL3NlbnNvci9QSwECPwMUAAAICAAAACEAsn%2FwvrgAAABeAQAAHAAAAAAAAAAAAAAApIGBJAAAcGhvdG9ubGFiL3NlbnNvci9fX2luaXRfXy5weVBLAQI%2FAxQAAAgIAAAAIQBj0IPujAEAAOsCAAAcAAAAAAAAAAAAAACkgXMlAABwaG90b25sYWIvc2Vuc29yL2dlb21ldHJ5LnB5UEsBAj8DFAAACAgAAAAhAMD2ZigKBAAAOQoAAB8AAAAAAAAAAAAAAKSBOScAAHBob3RvbmxhYi9zZW5zb3IvcGVyZm9ybWFuY2UucHlQSwECPwMUAAAICAAAACEA8wNtaiUCAAD%2FAwAAEgAAAAAAAAAAAAAApIGAKwAAcGhvdG9ubGFiL3VuaXRzLnB5UEsBAj8DFAAACAAAAAAhAAAAAAAAAAAAAAAAABoAAAAAAAAAAAAAAO1B1S0AAHBob3RvbmxhYi0wLjEuMC5kaXN0LWluZm8vUEsBAj8DFAAACAgAAAAhAOl6sF5RAAAAUAAAAB8AAAAAAAAAAAAAAKSBDS4AAHBob3RvbmxhYi0wLjEuMC5kaXN0LWluZm8vV0hFRUxQSwECPwMUAAAICAAAACEAqM1iAw0IAABzEQAAIgAAAAAAAAAAAAAApIGbLgAAcGhvdG9ubGFiLTAuMS4wLmRpc3QtaW5mby9NRVRBREFUQVBLAQI%2FAxQAAAgIAAAAIQAf4xPQLwQAAIoHAAAgAAAAAAAAAAAAAACkgeg2AABwaG90b25sYWItMC4xLjAuZGlzdC1pbmZvL1JFQ09SRFBLBQYAAAAAIwAjAOsJAABVOwAAAAA%3D%22%2C%20%22refs%22%3A%20%7B%22emva1288%3A2021%22%3A%20%7B%22cite%22%3A%20%22EMVA%201288%2C%202021%22%2C%20%22full%22%3A%20%22European%20Machine%20Vision%20Association%20(2021).%20*EMVA%20Standard%201288%3A%20Standard%20for%20Characterization%20of%20Image%20Sensors%20and%20Cameras%2C%20Release%204.0%20Linear*.%20European%20Machine%20Vision%20Association.%22%7D%2C%20%22janesick%3A2007%22%3A%20%7B%22cite%22%3A%20%22Janesick%2C%202007%22%2C%20%22full%22%3A%20%22Janesick%2C%20James%20R.%20(2007).%20*Photon%20Transfer%3A%20DN%20%5C%5Cu2192%20%5C%5Cu03bb*.%20SPIE%20Press.%22%7D%2C%20%22smith%3A2008%22%3A%20%7B%22cite%22%3A%20%22Smith%2C%202008%22%2C%20%22full%22%3A%20%22Smith%2C%20Warren%20J.%20(2008).%20*Modern%20Optical%20Engineering%3A%20The%20Design%20of%20Optical%20Systems*.%204th%20ed.%20McGraw-Hill.%22%7D%2C%20%22hecht%3A2017%22%3A%20%7B%22cite%22%3A%20%22Hecht%2C%202017%22%2C%20%22full%22%3A%20%22Hecht%2C%20Eugene%20(2017).%20*Optics*.%205th%20ed.%20Pearson.%22%7D%7D%7D')%0A%0A%20%20%20%20if%20sys.platform%20%3D%3D%20%22emscripten%22%3A%0A%20%20%20%20%20%20%20%20import%20micropip%0A%0A%20%20%20%20%20%20%20%20_whl%20%3D%20Path(%22%2Ftmp%22)%20%2F%20EMBEDDED%5B%22wheel%22%5D%0A%20%20%20%20%20%20%20%20_whl.write_bytes(base64.b64decode(EMBEDDED%5B%22wheel_b64%22%5D))%0A%20%20%20%20%20%20%20%20await%20micropip.install(f%22emfs%3A%7B_whl%7D%22)%0A%20%20%20%20%20%20%20%20refs%20%3D%20EMBEDDED%5B%22refs%22%5D%0A%20%20%20%20else%3A%0A%20%20%20%20%20%20%20%20from%20photonlab.nb%20import%20reference_texts%0A%0A%20%20%20%20%20%20%20%20_repo%20%3D%20next(p%20for%20p%20in%20Path(mo.notebook_dir()).resolve().parents%20if%20(p%20%2F%20%22refs.bib%22).exists())%0A%20%20%20%20%20%20%20%20refs%20%3D%20reference_texts(_repo%20%2F%20%22refs.bib%22)%0A%20%20%20%20photonlab_loaded%20%3D%20True%0A%20%20%20%20return%20mo%2C%20photonlab_loaded%2C%20refs%0A%0A%0Aif%20__name__%20%3D%3D%20%22__main__%22%3A%0A%20%20%20%20app.run()%0A