SlideShare una empresa de Scribd logo
1 de 34
Descargar para leer sin conexión
Satellite	Image	Segmentation	
for	
Photovoltaic	Potential	Estimation
Python	based	Google	Cloud	Platform	project
Johannes	Oos
oosjoh@gmail.com
1.Purpose	and	Input	Data	
2.Computer	Vision/	Models
3.Auto-ML
4.Summary
1.Purpose	and Input	Data	
2.Computer	Vision/	Models
3.Auto-ML
4.Summary
Purpose
Enter	address
Estimation	of	
system	key	
parameters:	
Size,	energy	yield
Address Size,	tilt,	
azimuth
climate	
data
Solar	
Simulation
Size,
Energy	
Data	Flow
Geo-location
lat /	lng
Datasource – Images	from	googlemaps
Geocoding	API
• To	translate	an	address	to	geo-coordinates
• Clean	and	Python	friendly	
• pip	install	-U	googlemaps
Maps	Static	API
• To	get	the	satelite image	via	http	request
• URL	parsing	with	urllib
Geocoding	API
Maps	Static	API	- request
Maps	Static	API	- image
1.Purpose	and	Input	Data	
2.Computer	Vision/	Models
3.Auto-ML
4.Summary
Image	Analysis
Image	Analysis
Classification Detection Segmentation
90%
http://cocodataset.org/#explore
Model Family Task Input Output Technology Model
1 Segmentation Create	Roof	Segments,
Determine	size	of	
roofs
Satellite	Image Roof Box,	
Segments
Keras
TensorFlow
Mask	R-CNN
2 Classification	 Decide	if	roof	is	flat	or	
tilted
Roof	Box	Image Label AutoML GCP
3 Classification Get	Tilt	angle Roof	Box	Image	of	
tilted roof
Label AutoML GCP
4 Classification	 Get	Azimuth	angle Roof	Box	Image	of	
tilted roof
Label AutoML GCP
Overview	– Used	Models
Model Family Task Input Output Technology Model
1 Segmentation Create	Roof	Segments,
Determine	size	of	
roofs
Satellite	Image Roof Box,	
Segments
Keras
TensorFlow
Mask	R-CNN
2 Classification	 Decide	if	roof	is	flat	or	
tilted
Roof	Box	Image Label AutoML GCP
3 Classification Get	Tilt	angle Roof	Box	Image	of	
tilted roof
Label AutoML GCP
4 Classification	 Get	Azimuth	angle Roof	Box	Image	of	
tilted roof
Label AutoML GCP
Overview	– Used	Models
Mask	R-CNN
• https://arxiv.org/abs/1703.06870
• https://github.com/matterport/Mask_RCNN
• A	well	documented,	easy	to	use	implementation	for	image	
segmentation	tasks
• Implemented	with	Python,	Tensorflow,	Keras
Mask	R-CNN
Blue	Bounding Box:	
Starting top	Left [	79,	15,	164,	68]
Pixels in	the Blue	Mask:	
==	>		2842	px
@Zoom	Level	19:	
Distance			è 0.2986	m	/	pixel
Area										è 11	pixel /	sqm
Blue	roof =	260	sqm
The	content of the Bounding Box
Is	used	for	classification
Model Family Task Input Output Technology Model
1 Segmentation Create	Roof	Segments,
Determine	size	of	
roofs
Satellite	Image Roof Box,	
Segments
Keras
TensorFlow
Mask	R-CNN
2 Classification	 Decide	if	roof	is	flat	or	
tilted
Roof	Box	Image Label AutoML GCP
3 Classification Get	Tilt	angle Roof	Box	Image	of	
tilted roof
Label AutoML GCP
4 Classification	 Get	Azimuth	angle Roof	Box	Image	of	
tilted roof
Label AutoML GCP
Overview	– Used	Models
Satellite	images	
:		“flat”
:		“tilted”
Label	data	set
GCP	AutoML
Customized	
Classifier
91%	Tilted
Pre-trained	Classifier
Retraining	a	Classifier
Choice	of	Classifier
1.Purpose	and	Input	Data	
2.Computer	Vision/	Models
3.Auto-ML
4.Summary
ML	APIs MLAutoML
Train	high-quality	custom	machine	learning	models	
with	minimum	effort	and	machine	learning	expertise
AutoML on	GCP
Data
Scientist
App	
developer
Vision
Natural	
Language
Translate
AutoML on	GCP
AutoML Admin		Prerequisites
• Set	Up	GCP	Account
• Enable	Billing	
• Create	GCP	Project
• Authorization
• Pip	install	API
• Enable	API	in	project
Admin	Prerequisites	– Set	Credentials
DefaultCredentialsError:	Could	not	automatically	determine	
credentials.	Please	set	GOOGLE_APPLICATION_CREDENTIALS
Admin	Prerequisites	– Enable	API
PermissionDenied:	403	[…]	API	has	not	been	used	in	project	
[…]	before	or	it	is	disabled
Train	– Uploading	and	Labelling	the	data
Train	and	Evaluate	the	models
Train	– Keeping	track	of	the	models
AutoML - Using	the	ready	model	(1/2)
Image
Type:	google.cloud.automl_v1beta1.
types.PredictResponse
payload {		classification {			
score:	 0.7124295830726624		}		
display_name:	"roof_20"}
res.payload[0].display_name
Personalized	Classifier	
Hosted	on	GCP
Response	
Object
Timing:	1	- 3 Seconds
AutoML - Using	the	ready	model	(2/2)
1.Purpose	and	Input	Data	
2.Computer	Vision/	Models
3.Auto-ML
4.Summary
Bringing	it	all	together
Size:	260	sqm
”Address”
”lat /	lng”
Irradiation Profile
30	kWp @
950	kWh/kWp/year	Simulation
Type:	Flat
Thank	you!
Using	AutoML
1. Very	comfortable	data	and	model	management
2. Python	friendly	usage
3. Still	in	Beta	mode
4. Good	for	Classification
5. Can’t	be	optimized	for	speed

Más contenido relacionado

La actualidad más candente

VERIFICATION_&_VALIDATION_OF_A_SEMANTIC_IMAGE_TAGGING_FRAMEWORK_VIA_GENERATIO...
VERIFICATION_&_VALIDATION_OF_A_SEMANTIC_IMAGE_TAGGING_FRAMEWORK_VIA_GENERATIO...VERIFICATION_&_VALIDATION_OF_A_SEMANTIC_IMAGE_TAGGING_FRAMEWORK_VIA_GENERATIO...
VERIFICATION_&_VALIDATION_OF_A_SEMANTIC_IMAGE_TAGGING_FRAMEWORK_VIA_GENERATIO...
grssieee
 

La actualidad más candente (20)

13 helioscope pvpmc 2017v4
13 helioscope pvpmc 2017v413 helioscope pvpmc 2017v4
13 helioscope pvpmc 2017v4
 
19 winter towards_an_energy-based_parameter_for_photovoltaic_classification
19 winter towards_an_energy-based_parameter_for_photovoltaic_classification19 winter towards_an_energy-based_parameter_for_photovoltaic_classification
19 winter towards_an_energy-based_parameter_for_photovoltaic_classification
 
53 aron p_dobos_recent_and_planned_improvements_to_the_system_advisor_model_sam
53 aron p_dobos_recent_and_planned_improvements_to_the_system_advisor_model_sam53 aron p_dobos_recent_and_planned_improvements_to_the_system_advisor_model_sam
53 aron p_dobos_recent_and_planned_improvements_to_the_system_advisor_model_sam
 
55 reinders performance_modelling_of_pv_systems_in_a_virtual_environment
55 reinders performance_modelling_of_pv_systems_in_a_virtual_environment55 reinders performance_modelling_of_pv_systems_in_a_virtual_environment
55 reinders performance_modelling_of_pv_systems_in_a_virtual_environment
 
Design Optimization using the Latest Features in HelioScope
Design Optimization using the Latest Features in HelioScopeDesign Optimization using the Latest Features in HelioScope
Design Optimization using the Latest Features in HelioScope
 
GRC - CANARIE Green IT 2009
GRC - CANARIE Green IT 2009GRC - CANARIE Green IT 2009
GRC - CANARIE Green IT 2009
 
33 freeman modelling_energy_losses_due_to_snow_on_pv_systems
33 freeman modelling_energy_losses_due_to_snow_on_pv_systems33 freeman modelling_energy_losses_due_to_snow_on_pv_systems
33 freeman modelling_energy_losses_due_to_snow_on_pv_systems
 
24 mavromatakis vignola_spectral_corrections_for_pv_performance_modelling
24 mavromatakis vignola_spectral_corrections_for_pv_performance_modelling24 mavromatakis vignola_spectral_corrections_for_pv_performance_modelling
24 mavromatakis vignola_spectral_corrections_for_pv_performance_modelling
 
11 schroedter homscheidt_satellite_and_camera
11 schroedter homscheidt_satellite_and_camera11 schroedter homscheidt_satellite_and_camera
11 schroedter homscheidt_satellite_and_camera
 
meteodynWT meso coupling downscaling regional planing
meteodynWT meso coupling downscaling regional planingmeteodynWT meso coupling downscaling regional planing
meteodynWT meso coupling downscaling regional planing
 
16 lorenz local_and_regional_pv_power
16 lorenz local_and_regional_pv_power16 lorenz local_and_regional_pv_power
16 lorenz local_and_regional_pv_power
 
Emotion recognition-from-multichannel-eeg-signal-using-2 d
Emotion recognition-from-multichannel-eeg-signal-using-2 dEmotion recognition-from-multichannel-eeg-signal-using-2 d
Emotion recognition-from-multichannel-eeg-signal-using-2 d
 
2018 China Gas Power Map
2018 China Gas Power Map2018 China Gas Power Map
2018 China Gas Power Map
 
66 ueda system_performance_and_degradation_analysis_of_different_pv_technologies
66 ueda system_performance_and_degradation_analysis_of_different_pv_technologies66 ueda system_performance_and_degradation_analysis_of_different_pv_technologies
66 ueda system_performance_and_degradation_analysis_of_different_pv_technologies
 
EMPLOYING MULTI CORE ARCHITECTURE TO OPTIMIZE ON PERFORMANCE, FOR APPROACH IN...
EMPLOYING MULTI CORE ARCHITECTURE TO OPTIMIZE ON PERFORMANCE, FOR APPROACH IN...EMPLOYING MULTI CORE ARCHITECTURE TO OPTIMIZE ON PERFORMANCE, FOR APPROACH IN...
EMPLOYING MULTI CORE ARCHITECTURE TO OPTIMIZE ON PERFORMANCE, FOR APPROACH IN...
 
VERIFICATION_&_VALIDATION_OF_A_SEMANTIC_IMAGE_TAGGING_FRAMEWORK_VIA_GENERATIO...
VERIFICATION_&_VALIDATION_OF_A_SEMANTIC_IMAGE_TAGGING_FRAMEWORK_VIA_GENERATIO...VERIFICATION_&_VALIDATION_OF_A_SEMANTIC_IMAGE_TAGGING_FRAMEWORK_VIA_GENERATIO...
VERIFICATION_&_VALIDATION_OF_A_SEMANTIC_IMAGE_TAGGING_FRAMEWORK_VIA_GENERATIO...
 
03 solargis uncertainty_albuquerque_pvp_mws_2017-05_final
03 solargis uncertainty_albuquerque_pvp_mws_2017-05_final03 solargis uncertainty_albuquerque_pvp_mws_2017-05_final
03 solargis uncertainty_albuquerque_pvp_mws_2017-05_final
 
51 b wittmer_latest_features_of_p_vsyst
51 b wittmer_latest_features_of_p_vsyst51 b wittmer_latest_features_of_p_vsyst
51 b wittmer_latest_features_of_p_vsyst
 
Secure power grid simulation on cloud
Secure power grid simulation on cloudSecure power grid simulation on cloud
Secure power grid simulation on cloud
 
05 2017 05-04-clear sky models g-kimball
05 2017 05-04-clear sky models g-kimball05 2017 05-04-clear sky models g-kimball
05 2017 05-04-clear sky models g-kimball
 

Similar a PyConDE18

Apache Hadoop India Summit 2011 Keynote talk "Scaling Hadoop Applications" by...
Apache Hadoop India Summit 2011 Keynote talk "Scaling Hadoop Applications" by...Apache Hadoop India Summit 2011 Keynote talk "Scaling Hadoop Applications" by...
Apache Hadoop India Summit 2011 Keynote talk "Scaling Hadoop Applications" by...
Yahoo Developer Network
 

Similar a PyConDE18 (20)

DSM Extraction from Pleiades Images using Micmac
DSM Extraction from Pleiades Images using MicmacDSM Extraction from Pleiades Images using Micmac
DSM Extraction from Pleiades Images using Micmac
 
Analytics Zoo: Building Analytics and AI Pipeline for Apache Spark and BigDL ...
Analytics Zoo: Building Analytics and AI Pipeline for Apache Spark and BigDL ...Analytics Zoo: Building Analytics and AI Pipeline for Apache Spark and BigDL ...
Analytics Zoo: Building Analytics and AI Pipeline for Apache Spark and BigDL ...
 
Introduction to mago3D, an Open Source Based Digital Twin Platform
Introduction to mago3D, an Open Source Based Digital Twin PlatformIntroduction to mago3D, an Open Source Based Digital Twin Platform
Introduction to mago3D, an Open Source Based Digital Twin Platform
 
Scaling hadoopapplications
Scaling hadoopapplicationsScaling hadoopapplications
Scaling hadoopapplications
 
ei2106-submit-opt-415
ei2106-submit-opt-415ei2106-submit-opt-415
ei2106-submit-opt-415
 
Remote Sensing Field Camp 2016
Remote Sensing Field Camp 2016 Remote Sensing Field Camp 2016
Remote Sensing Field Camp 2016
 
Apache Hadoop India Summit 2011 Keynote talk "Scaling Hadoop Applications" by...
Apache Hadoop India Summit 2011 Keynote talk "Scaling Hadoop Applications" by...Apache Hadoop India Summit 2011 Keynote talk "Scaling Hadoop Applications" by...
Apache Hadoop India Summit 2011 Keynote talk "Scaling Hadoop Applications" by...
 
FME User Stories from Around the World
FME User Stories from Around the WorldFME User Stories from Around the World
FME User Stories from Around the World
 
Project Matsu: Elastic Clouds for Disaster Relief
Project Matsu: Elastic Clouds for Disaster ReliefProject Matsu: Elastic Clouds for Disaster Relief
Project Matsu: Elastic Clouds for Disaster Relief
 
IRJET - A Survey Paper on Efficient Object Detection and Matching using F...
IRJET -  	  A Survey Paper on Efficient Object Detection and Matching using F...IRJET -  	  A Survey Paper on Efficient Object Detection and Matching using F...
IRJET - A Survey Paper on Efficient Object Detection and Matching using F...
 
Spar 2010 Presetation
Spar 2010 PresetationSpar 2010 Presetation
Spar 2010 Presetation
 
March 2011 HUG: Scaling Hadoop
March 2011 HUG: Scaling HadoopMarch 2011 HUG: Scaling Hadoop
March 2011 HUG: Scaling Hadoop
 
thesis-presentation
thesis-presentationthesis-presentation
thesis-presentation
 
OpenStreetMap in 3D - current developments
OpenStreetMap in 3D - current developmentsOpenStreetMap in 3D - current developments
OpenStreetMap in 3D - current developments
 
도시건축설계와 오픈소스 기반 GIS
도시건축설계와 오픈소스 기반 GIS도시건축설계와 오픈소스 기반 GIS
도시건축설계와 오픈소스 기반 GIS
 
도시 설계와 GIS 기술의 관계
도시 설계와 GIS 기술의 관계도시 설계와 GIS 기술의 관계
도시 설계와 GIS 기술의 관계
 
Electrolux meetup
Electrolux meetupElectrolux meetup
Electrolux meetup
 
Scalable AutoML for Time Series Forecasting using Ray
Scalable AutoML for Time Series Forecasting using RayScalable AutoML for Time Series Forecasting using Ray
Scalable AutoML for Time Series Forecasting using Ray
 
Current State of mago3D, an Open Source Based Digital Twin Platform
Current State of mago3D, an Open Source Based Digital Twin PlatformCurrent State of mago3D, an Open Source Based Digital Twin Platform
Current State of mago3D, an Open Source Based Digital Twin Platform
 
DSM Extraction from Pleiades Images using MICMAC
DSM Extraction from Pleiades Images using MICMAC DSM Extraction from Pleiades Images using MICMAC
DSM Extraction from Pleiades Images using MICMAC
 

Último

Structural Analysis and Design of Foundations: A Comprehensive Handbook for S...
Structural Analysis and Design of Foundations: A Comprehensive Handbook for S...Structural Analysis and Design of Foundations: A Comprehensive Handbook for S...
Structural Analysis and Design of Foundations: A Comprehensive Handbook for S...
Dr.Costas Sachpazis
 
Call Now ≽ 9953056974 ≼🔝 Call Girls In New Ashok Nagar ≼🔝 Delhi door step de...
Call Now ≽ 9953056974 ≼🔝 Call Girls In New Ashok Nagar  ≼🔝 Delhi door step de...Call Now ≽ 9953056974 ≼🔝 Call Girls In New Ashok Nagar  ≼🔝 Delhi door step de...
Call Now ≽ 9953056974 ≼🔝 Call Girls In New Ashok Nagar ≼🔝 Delhi door step de...
9953056974 Low Rate Call Girls In Saket, Delhi NCR
 

Último (20)

Booking open Available Pune Call Girls Koregaon Park 6297143586 Call Hot Ind...
Booking open Available Pune Call Girls Koregaon Park  6297143586 Call Hot Ind...Booking open Available Pune Call Girls Koregaon Park  6297143586 Call Hot Ind...
Booking open Available Pune Call Girls Koregaon Park 6297143586 Call Hot Ind...
 
Double rodded leveling 1 pdf activity 01
Double rodded leveling 1 pdf activity 01Double rodded leveling 1 pdf activity 01
Double rodded leveling 1 pdf activity 01
 
Structural Analysis and Design of Foundations: A Comprehensive Handbook for S...
Structural Analysis and Design of Foundations: A Comprehensive Handbook for S...Structural Analysis and Design of Foundations: A Comprehensive Handbook for S...
Structural Analysis and Design of Foundations: A Comprehensive Handbook for S...
 
ONLINE FOOD ORDER SYSTEM PROJECT REPORT.pdf
ONLINE FOOD ORDER SYSTEM PROJECT REPORT.pdfONLINE FOOD ORDER SYSTEM PROJECT REPORT.pdf
ONLINE FOOD ORDER SYSTEM PROJECT REPORT.pdf
 
(INDIRA) Call Girl Bhosari Call Now 8617697112 Bhosari Escorts 24x7
(INDIRA) Call Girl Bhosari Call Now 8617697112 Bhosari Escorts 24x7(INDIRA) Call Girl Bhosari Call Now 8617697112 Bhosari Escorts 24x7
(INDIRA) Call Girl Bhosari Call Now 8617697112 Bhosari Escorts 24x7
 
Intze Overhead Water Tank Design by Working Stress - IS Method.pdf
Intze Overhead Water Tank  Design by Working Stress - IS Method.pdfIntze Overhead Water Tank  Design by Working Stress - IS Method.pdf
Intze Overhead Water Tank Design by Working Stress - IS Method.pdf
 
PVC VS. FIBERGLASS (FRP) GRAVITY SEWER - UNI BELL
PVC VS. FIBERGLASS (FRP) GRAVITY SEWER - UNI BELLPVC VS. FIBERGLASS (FRP) GRAVITY SEWER - UNI BELL
PVC VS. FIBERGLASS (FRP) GRAVITY SEWER - UNI BELL
 
Unit 1 - Soil Classification and Compaction.pdf
Unit 1 - Soil Classification and Compaction.pdfUnit 1 - Soil Classification and Compaction.pdf
Unit 1 - Soil Classification and Compaction.pdf
 
(INDIRA) Call Girl Aurangabad Call Now 8617697112 Aurangabad Escorts 24x7
(INDIRA) Call Girl Aurangabad Call Now 8617697112 Aurangabad Escorts 24x7(INDIRA) Call Girl Aurangabad Call Now 8617697112 Aurangabad Escorts 24x7
(INDIRA) Call Girl Aurangabad Call Now 8617697112 Aurangabad Escorts 24x7
 
Call Now ≽ 9953056974 ≼🔝 Call Girls In New Ashok Nagar ≼🔝 Delhi door step de...
Call Now ≽ 9953056974 ≼🔝 Call Girls In New Ashok Nagar  ≼🔝 Delhi door step de...Call Now ≽ 9953056974 ≼🔝 Call Girls In New Ashok Nagar  ≼🔝 Delhi door step de...
Call Now ≽ 9953056974 ≼🔝 Call Girls In New Ashok Nagar ≼🔝 Delhi door step de...
 
Call Girls Walvekar Nagar Call Me 7737669865 Budget Friendly No Advance Booking
Call Girls Walvekar Nagar Call Me 7737669865 Budget Friendly No Advance BookingCall Girls Walvekar Nagar Call Me 7737669865 Budget Friendly No Advance Booking
Call Girls Walvekar Nagar Call Me 7737669865 Budget Friendly No Advance Booking
 
Generative AI or GenAI technology based PPT
Generative AI or GenAI technology based PPTGenerative AI or GenAI technology based PPT
Generative AI or GenAI technology based PPT
 
University management System project report..pdf
University management System project report..pdfUniversity management System project report..pdf
University management System project report..pdf
 
data_management_and _data_science_cheat_sheet.pdf
data_management_and _data_science_cheat_sheet.pdfdata_management_and _data_science_cheat_sheet.pdf
data_management_and _data_science_cheat_sheet.pdf
 
BSides Seattle 2024 - Stopping Ethan Hunt From Taking Your Data.pptx
BSides Seattle 2024 - Stopping Ethan Hunt From Taking Your Data.pptxBSides Seattle 2024 - Stopping Ethan Hunt From Taking Your Data.pptx
BSides Seattle 2024 - Stopping Ethan Hunt From Taking Your Data.pptx
 
Thermal Engineering-R & A / C - unit - V
Thermal Engineering-R & A / C - unit - VThermal Engineering-R & A / C - unit - V
Thermal Engineering-R & A / C - unit - V
 
Roadmap to Membership of RICS - Pathways and Routes
Roadmap to Membership of RICS - Pathways and RoutesRoadmap to Membership of RICS - Pathways and Routes
Roadmap to Membership of RICS - Pathways and Routes
 
Booking open Available Pune Call Girls Pargaon 6297143586 Call Hot Indian Gi...
Booking open Available Pune Call Girls Pargaon  6297143586 Call Hot Indian Gi...Booking open Available Pune Call Girls Pargaon  6297143586 Call Hot Indian Gi...
Booking open Available Pune Call Girls Pargaon 6297143586 Call Hot Indian Gi...
 
chapter 5.pptx: drainage and irrigation engineering
chapter 5.pptx: drainage and irrigation engineeringchapter 5.pptx: drainage and irrigation engineering
chapter 5.pptx: drainage and irrigation engineering
 
KubeKraft presentation @CloudNativeHooghly
KubeKraft presentation @CloudNativeHooghlyKubeKraft presentation @CloudNativeHooghly
KubeKraft presentation @CloudNativeHooghly
 

PyConDE18