LiLz uses computer vision to read gauges and dials where humans prefer not to tread – TechCrunch

No one wants to be the maintenance worker who has to hike through half a mile of damp hallways just to check the pressure gauge on a valve somewhere. LiLz makes it possible to keep an eye on such inconvenient physical interfaces remotely with a clever and practical application of machine learning.

The Japanese (specifically Okinawan) startup has been around for a little while — in fact our colleagues at TC JP have written them up. But despite the seemingly obvious value of its service, it hasn’t quite hit the big time yet. LiLz participated in CES as part of the country’s trade group, along with a bunch more companies listed here.

LiLz’s device looks a bit like a chubby tablet without a screen. It’s essentially a camera, light and processing and communications chips packed in with a big battery — enough power to last up to three years.

You mount the device so it can see the gauge or dial in question. After confirming picture and signal, you configure it in the app to interpret what it’s pointed at; it can read circular, semicircular and linear gauges, digital and rolling or analog digit displays, or things like colored warning lights. (The ML involved here is not trivial — I ran across this interesting article while looking into it.)

Once it’s set up, it’ll send readings live or at intervals to a central dashboard, or make them available via API so they can be queried or recorded elsewhere. The data goes out via LTE or Bluetooth.

Two Lilz cameras mounted near the ceiling of an industrial area.

Image Credits: LiLz

It’s a solution aimed directly at infrastructure and heavy industry, which often involve lots of legacy equipment located in hard-to-reach places: roofs, underground (but not too deep or the signal can’t penetrate), in labyrinthine factories and warehouses, etc.

Doing daily rounds of these dials is not just dull work for humans, but can be dangerous as well. Using a robot is another way to automate it, but doesn’t a network of IoT devices seem more practical than a quadrupedal bot trucking around constantly?

LiLz CTO Kuba Kolodziejczyk said that the company has been expanding quickly following its debut in 2020 and a $2.2 million Series A round in early 2021.

Screenshot of the Lilz interface showing readings from a dial being made by the software.

Image Credits: LiLz

“We grew from 240 cameras deployed at 34 locations by a few early users before June 2020, to 2000 cameras deployed at 320 locations for over 100 clients now, and expect to grow to 5000 cameras by end this year,” he wrote in an email to TechCrunch. “We now have a few clients using more than hundred cameras at single location.”

They started by targeting basic building management, but now have expanded to chemical and industrial plants, construction and manufacturing sites, public infrastructure and more due to customer demand, he added.

And the capabilities of the device are increasing as well. They’ve mainly been iterating on the software, making it able to be updated remotely, improving accuracy and resilience to disturbance, and adding data sharing and other features. There’s a new explosion-proof (!) version of the hardware as well. Maybe they should get into making phone cases as a side gig.

New products are on the way, one for “sound search” and another for counting of objects in the camera’s view, but both are at a very early stage now. And new capabilities like monitoring float and level gauges, among other things.

Meanwhile, they’ve come to CES to make some outreach to potential clients outside Japan. The show is looking like something of a bust, but maybe a few techs and managers will read this and think “wow, that would be handy.” You can find out more at the shiny new English website here.

Read more about CES 2022 on TechCrunch

No one wants to be the maintenance worker who has to hike through half a mile of damp hallways just to check the pressure gauge on a valve somewhere. LiLz makes it possible to keep an eye on such inconvenient physical interfaces remotely with a clever and practical application of machine learning.

The Japanese (specifically Okinawan) startup has been around for a little while — in fact our colleagues at TC JP have written them up. But despite the seemingly obvious value of its service, it hasn’t quite hit the big time yet. LiLz participated in CES as part of the country’s trade group, along with a bunch more companies listed here.

LiLz’s device looks a bit like a chubby tablet without a screen. It’s essentially a camera, light and processing and communications chips packed in with a big battery — enough power to last up to three years.

You mount the device so it can see the gauge or dial in question. After confirming picture and signal, you configure it in the app to interpret what it’s pointed at; it can read circular, semicircular and linear gauges, digital and rolling or analog digit displays, or things like colored warning lights. (The ML involved here is not trivial — I ran across this interesting article while looking into it.)

Once it’s set up, it’ll send readings live or at intervals to a central dashboard, or make them available via API so they can be queried or recorded elsewhere. The data goes out via LTE or Bluetooth.

Two Lilz cameras mounted near the ceiling of an industrial area.

Image Credits: LiLz

It’s a solution aimed directly at infrastructure and heavy industry, which often involve lots of legacy equipment located in hard-to-reach places: roofs, underground (but not too deep or the signal can’t penetrate), in labyrinthine factories and warehouses, etc.

Doing daily rounds of these dials is not just dull work for humans, but can be dangerous as well. Using a robot is another way to automate it, but doesn’t a network of IoT devices seem more practical than a quadrupedal bot trucking around constantly?

LiLz CTO Kuba Kolodziejczyk said that the company has been expanding quickly following its debut in 2020 and a $2.2 million Series A round in early 2021.

Screenshot of the Lilz interface showing readings from a dial being made by the software.

Image Credits: LiLz

“We grew from 240 cameras deployed at 34 locations by a few early users before June 2020, to 2000 cameras deployed at 320 locations for over 100 clients now, and expect to grow to 5000 cameras by end this year,” he wrote in an email to TechCrunch. “We now have a few clients using more than hundred cameras at single location.”

They started by targeting basic building management, but now have expanded to chemical and industrial plants, construction and manufacturing sites, public infrastructure and more due to customer demand, he added.

And the capabilities of the device are increasing as well. They’ve mainly been iterating on the software, making it able to be updated remotely, improving accuracy and resilience to disturbance, and adding data sharing and other features. There’s a new explosion-proof (!) version of the hardware as well. Maybe they should get into making phone cases as a side gig.

New products are on the way, one for “sound search” and another for counting of objects in the camera’s view, but both are at a very early stage now. And new capabilities like monitoring float and level gauges, among other things.

Meanwhile, they’ve come to CES to make some outreach to potential clients outside Japan. The show is looking like something of a bust, but maybe a few techs and managers will read this and think “wow, that would be handy.” You can find out more at the shiny new English website here.

Read more about CES 2022 on TechCrunch

More from author

Related posts

Advertismentspot_img

Latest posts

Apple AirPods Pro (2nd Generation) Will Feature Hearing Aid Function, Heart Rate Detection: Report

Apple AirPods Pro (2nd Generation) is reportedly going to launch with a long list of features like heart rate detection, hearing aid function, and...

Fossil Gen 6 Hybrid Smartwatch to Launch on June 27 With Up to 2 Weeks of Battery Life

Fossil Gen 6 smartwatch range was launched in August last year with a Qualcomm Snapdragon 4100+ SoC and SpO2 sensor. Now, Fossil is preparing...

Noise Expects to Make Its Revenue Double to Rs. 2,000 Crore in Current Fiscal Year

Wearable and hearable device maker Noise expects to more than double its revenue in the current financial year to Rs. 2,000 crore, a top...
[tdn_block_newsletter_subscribe title_text="Want to stay up to date with the latest news? " description="V2UlMjB3b3VsZCUyMGxvdmUlMjB0byUyMGhlYXIlMjBmcm9tJTIweW91ISUyMFBsZWFzZSUyMGZpbGwlMjBpbiUyMHlvdXIlMjBkZXRhaWxzJTIwYW5kJTIwd2UlMjB3aWxsJTIwc3RheSUyMGluJTIwdG91Y2guJTIwSXQncyUyMHRoYXQlMjBzaW1wbGUh" input_placeholder="Email address" btn_text="Subscribe" tds_newsletter2-image="8" tds_newsletter2-image_bg_color="#c3ecff" tds_newsletter3-input_bar_display="row" tds_newsletter4-image="9" tds_newsletter4-image_bg_color="#fffbcf" tds_newsletter4-btn_bg_color="#f3b700" tds_newsletter4-check_accent="#f3b700" tds_newsletter5-tdicon="tdc-font-fa tdc-font-fa-envelope-o" tds_newsletter5-btn_bg_color="#000000" tds_newsletter5-btn_bg_color_hover="#4db2ec" tds_newsletter5-check_accent="#000000" tds_newsletter6-input_bar_display="row" tds_newsletter6-btn_bg_color="#da1414" tds_newsletter6-check_accent="#da1414" tds_newsletter7-image="10" tds_newsletter7-btn_bg_color="#1c69ad" tds_newsletter7-check_accent="#1c69ad" tds_newsletter7-f_title_font_size="20" tds_newsletter7-f_title_font_line_height="28px" tds_newsletter8-input_bar_display="row" tds_newsletter8-btn_bg_color="#00649e" tds_newsletter8-btn_bg_color_hover="#21709e" tds_newsletter8-check_accent="#00649e" embedded_form_code="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" tds_newsletter="tds_newsletter1" tds_newsletter1-input_bar_display="" tds_newsletter1-input_border_size="0" tds_newsletter1-title_color="#172842" tds_newsletter1-description_color="#90a0af" tds_newsletter1-disclaimer_color="#90a0af" tds_newsletter1-disclaimer2_color="#90a0af" tds_newsletter1-input_text_color="#90a0af" tds_newsletter1-input_placeholder_color="#bcccd6" tds_newsletter1-input_bg_color="#ffffff" tds_newsletter1-input_border_color="rgba(255,255,255,0)" tds_newsletter1-input_border_color_active="rgba(255,255,255,0)" tds_newsletter1-f_title_font_family="394" tds_newsletter1-f_title_font_size="eyJhbGwiOiI0MiIsImxhbmRzY2FwZSI6IjM2IiwicG9ydHJhaXQiOiIzMCIsInBob25lIjoiMzAifQ==" tds_newsletter1-f_title_font_line_height="1.2" tds_newsletter1-f_title_font_spacing="-1" tds_newsletter1-f_descr_font_family="638" tds_newsletter1-f_descr_font_size="eyJhbGwiOiIxOCIsImxhbmRzY2FwZSI6IjE1IiwicG9ydHJhaXQiOiIxNCIsInBob25lIjoiMTQifQ==" tds_newsletter1-f_descr_font_line_height="1.6" tds_newsletter1-f_descr_font_weight="700" content_align_horizontal="content-horiz-center" tdc_css="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" tds_newsletter1-f_disclaimer_font_family="394" tds_newsletter1-f_disclaimer2_font_family="394" tds_newsletter1-f_input_font_family="394" tds_newsletter1-f_input_font_line_height="3" tds_newsletter1-f_input_font_size="eyJhbGwiOiIxNiIsInBvcnRyYWl0IjoiMTQiLCJwaG9uZSI6IjE0In0=" tds_newsletter1-f_btn_font_family="394" tds_newsletter1-f_btn_font_transform="uppercase" tds_newsletter1-f_btn_font_weight="700" tds_newsletter1-btn_bg_color="#e2687e" tds_newsletter1-btn_bg_color_hover="#172842" tds_newsletter1-f_input_font_weight="" tds_newsletter1-f_title_font_weight="800"]