the sensor itself has a physical dimension now the sensors in maybe in dlsr cameras or in other compact cameras might be for example 35 mm is the biggest that comes back to the idea of when we had film and then you might have the micro 4/3 uh system for example and then down in a camera they're called like 1 over 2.5 in i won't go into this now an inch means 16 mm of course it's not it's again it's a hangover from an old days of lenses and cameras but basically it's like 1 over two half an inch 1 over 2.5 1 over three a third of an inch and that shows you that it is quite small now the smaller the sensor the less light it can capture that's obvious but this is where megapixels come in if you have many many many MEAP pixels packed into a very very tiny sensor then each of those pixels can only pick up a fraction of that light when in fact the size of the sensor divided by the number of pixels there are now if you have less megapixels and a bigger individual pixel size then each individual pixel is able to capture more light if you go too big without increasing the sensor size then the camera will start to produce more noisy pictures and it won't have the uh the the resolution in terms of light and that's when people start to talk about it's got a 1.4 or a 1.7 micrometer uh pixel size so this is an interesting thing that's that people manufacturers are now talking about how big are those individual pixels now if you go up to a big camera like a dlsr or a Micro 4 third system or something like that and of course you got a much bigger sense and there if you've got a 20 megapixel or a 21 megapixel camera that's much of a different game than it is if you have uh 21 megapixels in a small camera on a small sensor and once that light has been captured by the camera it's passed over to the ISP the image signal processor and companies like arm maker an image signal processor for their customers Qualcomm have their own image signal processor in their Snapdragon chips and basically it's responsible for taking all that data and producing the jpeg file in some cameras it can produce a raw file it has to apply the color profiles it has to do any postprocessing like HDR and that is all handled inside of the is SP and a couple of other quick things to talk about one is autofocus the picture needs to be focused and there are different methods for doing uh autofocus including a dual pixel method including phase detection
How smartphone cameras work β Gary explains
"WEBVTTKind: captionsLanguage: enhello there I'm Gary Sims from Android authority now I'm sure you've had a conversation with somebody about the camera in your smartphone and they've asked you well how many megapixels is it now while that's an important thing to ask it's not the only thing we should be asking about our smartphone cameras so the question before us today is how do smartphone cameras work and what are the important attributes well let me explain so a quick recap on how cameras work there's an amount of light in a scene that you want to take a photograph of whether it's indoors or outdoors and that light enters into your camera via a lens behind the lens there is an aperture which is basically a hole that lets the light through to the sensor that's behind it and it can be vary to be bigger or to be smaller and once it gets through to the sensor there is a way of exposing the sensor for a certain amount of time to the light if you expose it too much the picture will be all white if you don't expose enough the picture will be all black now in a dlsr or an SLR that would be with an actual physical shutter with a smartphone basically the sensor is switched on for a certain amount of time to capture and then switched off again and once all that data has been captured that light information it's then sent to an image signal processor an ISP which will take all that information and turn it into a JPEG photo according to various color profiles and it'll also apply maybe some postprocessing like HDR now there is an analogy to this which you'll find for example on the Cambridge in color photo tutorial website which talks about how if you want to capture some rain now the amount of rain that's pouring down out of the sky at any one moment is outside of your control however if you want to capture some water in a bucket there are certain things that you do have in your control for example the size of the bucket the bigger the bucket the more rain will enter into it you can also decide how long you leave the bucket outside and of course you are able to measure how much water you have captured and applying this idea to photograph the width the size of the bucket is the same as the aperture how much rain you're able to get in how much light it's able to get into the sensor the duration that you leave it out there is the same as the shutter speed and of course the iso tells you how much sensitivity there is how much the ability there is in capturing that water so let's start with the lens so when you read about lenses you will read a thing about the focal length and that's normally expressed in mm so on a dlsr for example this camera that I'm using to record with now has an 18 to 55 mm lens that's the focal length now on a wide camera a wide shot is somewhere like 18 20 23 and a zoomed in shot would be much more like 200 250 300 so the focal length basically defines how zoomed in you are on a particular thing and once the light has gone through the lens we arrive at the aperture now the larger the aperture the more light that's able to get in and having that large amount of light getting in also affects What's called the depth of field the bigger the aperture the narrower the depth of field if you want to get take some of those photos where the background is blurry and the subject is in crisp Focus then you need a wide aperture now you'll see that written on sometimes on the specifications about a phone an F 1.7 lens that tells you that's the widest that it can open now here's a strange thing about focal length the smaller the number number the bigger the hole it's inverse so 1.7 means a bigger hole and eight means a small hole now if you go over to the article that accompanies this uh particular video there's a whole bunch of uh details about how the focal length and how the aperture work together to give you those numbers but if you just want to know the basics the bigger the hole the more light they can get in but the shallower the depth of field and the smaller the number means the bigger the hole f1.7 big f1.5 big uh f8 means small now once the light has gone through the aperture it then gets to the sensor and there is a way of controlling how long the sensor is exposed to that light and that's using the shutter speed whether it be a physical shutter or an electronic shutter now basically if you're in a bright sunny day and you've got the aperture open wide then you only want the sensor to be exposed for just a moment to that amount of light but if you're in a low light situation or maybe if you've got the aperture smaller because you want a greater depth of field then you have to let the sensor be exposed for a longer amount of time now the problem with long exposures is we can't hold cameras very very still and that means you get camera Shake now of course there are ways around this you use a tripod that's one thing there's Optical image stabilization in some smartphones that's another thing but at the end of the day the longer the exposure time the long the slower the shutter speed then the more Shake there's going to be in your photos now that's not always a bad thing if you've seen the photos of kind of cut car lights that have been captured at night time the red streaming up the road then that's because they've left the shutter open for a long time with a small aperture and therefore only a little bit of light is getting in but it's able to trace those car Tails as they go through the city but if you'd have to do that on a tripod if you try to do it by hand then they'll all be shaky all over the place so a slow shut speed isn't necessarily bad but for a quick snap you need to be aware that you want a quick shutter speed and then of course the light gets through to the center sensor itself now the sensor can be tuned to have a certain amount of sensitivity and that's tuned in the same way as film used to be tuned with an ISO number ISO 100 ISO 400 ISO 800 and that basically means how sensitive it was that photographic payer to light and now how sensitive is the uh the sensor to light now basically sensitivity means how much is there a difference between one level of registered light and another if there is a big difference between the different color Shades then that means that there is going to be lots of crisp details however if there is a small distance then any kind of variation in the light will produce a variation in the color and what actually happens when you go way to the Other Extreme maybe you start to get up to uh 1,600 3200 and even higher then what you actually get is you get noise in the picture because the camera can't really tell the difference between one bit of light and another bit of light and it thinks they're two Shades of Gray two shades of red and your pictures start to look speckled they start to look dirty because the color isn't uniform so that's the danger of high ISO speed they introduce noise into the pictures and then the sensor itself has a physical Dimension now the sensors in maybe in dlsr cameras or in other compact cameras might be for example 35 mm is the biggest that comes back to the idea of when we had film and then you might have the micro 4/3 uh system for example and then down in a camera you they're called like 1 over 2.5 in I won't go into this now an inch means 16 mm of course it's not it's again it's a hangover from an old days of lenses and cameras but basically it's like 1 over two half an inch 1 over 2.5 1 over three a third of an inch and that shows you that it is quite small now the smaller the sensor the less light it can capture that's obvious but this is where megapixels come in if you have many many many MEAP pixels packed into a very very tiny sensor then each of those pixels can only pick up a fraction of that light when in fact the size of the sensor divided by the number of pixels there are now if you have less megapixels and a bigger individual pixel size then each individual pixel is able to capture more light if you go too big without increasing the sensor size then the camera will start to produce more noisy pictures and it won't have the uh the the resolution in terms of light and that's when people start to talk about it's got a 1.4 or a 1.7 micrometer uh pixel size so this is an interesting thing that's that people manufacturers are now talking about how big are those individual pixels now if you go up to a big camera like a dlsr or a Micro 4 third system or something like that and of course you got a much bigger sense and there if you've got a 20 megapixel or a 21 megapixel camera that's much of a different game than it is if you have uh 21 megapixels in a small camera on a small sensor and once that light has been captured by the camera it's passed over to the ISP the image signal processor and companies like arm maker an image signal processor for their customers Qualcomm have their own image signal processor in their Snapdragon chips and basically it's responsible for taking all that data and producing the jpeg file in some cameras it can produce a raw file it has to apply the color profiles it has to do any postprocessing like HDR and that is all handled inside of the is SP and a couple of other quick things to talk about one is autofocus the picture needs to be focused and there are different methods for doing uh autofocus including a dual pixel method including phase detection if you go over to the article that accompanies this video you'll see more details on how these different uh auto focus systems work and finally let's mention dual cameras if you look at a phone like the LG G6 what LG have done is they've put in lenses with two different focal lengths one with a wider one and with a more standard one and that means it's able to have options when it comes to taking photos if you look at a system like the Huawei P10 then that's using two sensors one monochrome and one color and the ISP is able to take the data from the monochrome sensor and really double up the amount of light that's coming into the picture particularly good for when you're doing things like HDR or you're in low light situations now I'd like to see more dual camera systems on smartphones I think it's a good way to compensate for the particularly small sensor size and the limitations that we have in having such small lenses I think dual cameras is the way to go however we noticed that Samsung didn't put that in the S8 this year so it'll be interesting to see what we get further down the road from companies like uh Samsung well I'm Gary Sims from Android authority I hope you enjoyed this video I hope you understood now some of the key attributes of smartphone cameras if you did please do give this video a thumbs up also don't forget to subscribe to Android authority don't forget to hit that Bell icon so that you get notifications when we publish a video and please do go over to Android authority.com because we are your source for all things Androidhello there I'm Gary Sims from Android authority now I'm sure you've had a conversation with somebody about the camera in your smartphone and they've asked you well how many megapixels is it now while that's an important thing to ask it's not the only thing we should be asking about our smartphone cameras so the question before us today is how do smartphone cameras work and what are the important attributes well let me explain so a quick recap on how cameras work there's an amount of light in a scene that you want to take a photograph of whether it's indoors or outdoors and that light enters into your camera via a lens behind the lens there is an aperture which is basically a hole that lets the light through to the sensor that's behind it and it can be vary to be bigger or to be smaller and once it gets through to the sensor there is a way of exposing the sensor for a certain amount of time to the light if you expose it too much the picture will be all white if you don't expose enough the picture will be all black now in a dlsr or an SLR that would be with an actual physical shutter with a smartphone basically the sensor is switched on for a certain amount of time to capture and then switched off again and once all that data has been captured that light information it's then sent to an image signal processor an ISP which will take all that information and turn it into a JPEG photo according to various color profiles and it'll also apply maybe some postprocessing like HDR now there is an analogy to this which you'll find for example on the Cambridge in color photo tutorial website which talks about how if you want to capture some rain now the amount of rain that's pouring down out of the sky at any one moment is outside of your control however if you want to capture some water in a bucket there are certain things that you do have in your control for example the size of the bucket the bigger the bucket the more rain will enter into it you can also decide how long you leave the bucket outside and of course you are able to measure how much water you have captured and applying this idea to photograph the width the size of the bucket is the same as the aperture how much rain you're able to get in how much light it's able to get into the sensor the duration that you leave it out there is the same as the shutter speed and of course the iso tells you how much sensitivity there is how much the ability there is in capturing that water so let's start with the lens so when you read about lenses you will read a thing about the focal length and that's normally expressed in mm so on a dlsr for example this camera that I'm using to record with now has an 18 to 55 mm lens that's the focal length now on a wide camera a wide shot is somewhere like 18 20 23 and a zoomed in shot would be much more like 200 250 300 so the focal length basically defines how zoomed in you are on a particular thing and once the light has gone through the lens we arrive at the aperture now the larger the aperture the more light that's able to get in and having that large amount of light getting in also affects What's called the depth of field the bigger the aperture the narrower the depth of field if you want to get take some of those photos where the background is blurry and the subject is in crisp Focus then you need a wide aperture now you'll see that written on sometimes on the specifications about a phone an F 1.7 lens that tells you that's the widest that it can open now here's a strange thing about focal length the smaller the number number the bigger the hole it's inverse so 1.7 means a bigger hole and eight means a small hole now if you go over to the article that accompanies this uh particular video there's a whole bunch of uh details about how the focal length and how the aperture work together to give you those numbers but if you just want to know the basics the bigger the hole the more light they can get in but the shallower the depth of field and the smaller the number means the bigger the hole f1.7 big f1.5 big uh f8 means small now once the light has gone through the aperture it then gets to the sensor and there is a way of controlling how long the sensor is exposed to that light and that's using the shutter speed whether it be a physical shutter or an electronic shutter now basically if you're in a bright sunny day and you've got the aperture open wide then you only want the sensor to be exposed for just a moment to that amount of light but if you're in a low light situation or maybe if you've got the aperture smaller because you want a greater depth of field then you have to let the sensor be exposed for a longer amount of time now the problem with long exposures is we can't hold cameras very very still and that means you get camera Shake now of course there are ways around this you use a tripod that's one thing there's Optical image stabilization in some smartphones that's another thing but at the end of the day the longer the exposure time the long the slower the shutter speed then the more Shake there's going to be in your photos now that's not always a bad thing if you've seen the photos of kind of cut car lights that have been captured at night time the red streaming up the road then that's because they've left the shutter open for a long time with a small aperture and therefore only a little bit of light is getting in but it's able to trace those car Tails as they go through the city but if you'd have to do that on a tripod if you try to do it by hand then they'll all be shaky all over the place so a slow shut speed isn't necessarily bad but for a quick snap you need to be aware that you want a quick shutter speed and then of course the light gets through to the center sensor itself now the sensor can be tuned to have a certain amount of sensitivity and that's tuned in the same way as film used to be tuned with an ISO number ISO 100 ISO 400 ISO 800 and that basically means how sensitive it was that photographic payer to light and now how sensitive is the uh the sensor to light now basically sensitivity means how much is there a difference between one level of registered light and another if there is a big difference between the different color Shades then that means that there is going to be lots of crisp details however if there is a small distance then any kind of variation in the light will produce a variation in the color and what actually happens when you go way to the Other Extreme maybe you start to get up to uh 1,600 3200 and even higher then what you actually get is you get noise in the picture because the camera can't really tell the difference between one bit of light and another bit of light and it thinks they're two Shades of Gray two shades of red and your pictures start to look speckled they start to look dirty because the color isn't uniform so that's the danger of high ISO speed they introduce noise into the pictures and then the sensor itself has a physical Dimension now the sensors in maybe in dlsr cameras or in other compact cameras might be for example 35 mm is the biggest that comes back to the idea of when we had film and then you might have the micro 4/3 uh system for example and then down in a camera you they're called like 1 over 2.5 in I won't go into this now an inch means 16 mm of course it's not it's again it's a hangover from an old days of lenses and cameras but basically it's like 1 over two half an inch 1 over 2.5 1 over three a third of an inch and that shows you that it is quite small now the smaller the sensor the less light it can capture that's obvious but this is where megapixels come in if you have many many many MEAP pixels packed into a very very tiny sensor then each of those pixels can only pick up a fraction of that light when in fact the size of the sensor divided by the number of pixels there are now if you have less megapixels and a bigger individual pixel size then each individual pixel is able to capture more light if you go too big without increasing the sensor size then the camera will start to produce more noisy pictures and it won't have the uh the the resolution in terms of light and that's when people start to talk about it's got a 1.4 or a 1.7 micrometer uh pixel size so this is an interesting thing that's that people manufacturers are now talking about how big are those individual pixels now if you go up to a big camera like a dlsr or a Micro 4 third system or something like that and of course you got a much bigger sense and there if you've got a 20 megapixel or a 21 megapixel camera that's much of a different game than it is if you have uh 21 megapixels in a small camera on a small sensor and once that light has been captured by the camera it's passed over to the ISP the image signal processor and companies like arm maker an image signal processor for their customers Qualcomm have their own image signal processor in their Snapdragon chips and basically it's responsible for taking all that data and producing the jpeg file in some cameras it can produce a raw file it has to apply the color profiles it has to do any postprocessing like HDR and that is all handled inside of the is SP and a couple of other quick things to talk about one is autofocus the picture needs to be focused and there are different methods for doing uh autofocus including a dual pixel method including phase detection if you go over to the article that accompanies this video you'll see more details on how these different uh auto focus systems work and finally let's mention dual cameras if you look at a phone like the LG G6 what LG have done is they've put in lenses with two different focal lengths one with a wider one and with a more standard one and that means it's able to have options when it comes to taking photos if you look at a system like the Huawei P10 then that's using two sensors one monochrome and one color and the ISP is able to take the data from the monochrome sensor and really double up the amount of light that's coming into the picture particularly good for when you're doing things like HDR or you're in low light situations now I'd like to see more dual camera systems on smartphones I think it's a good way to compensate for the particularly small sensor size and the limitations that we have in having such small lenses I think dual cameras is the way to go however we noticed that Samsung didn't put that in the S8 this year so it'll be interesting to see what we get further down the road from companies like uh Samsung well I'm Gary Sims from Android authority I hope you enjoyed this video I hope you understood now some of the key attributes of smartphone cameras if you did please do give this video a thumbs up also don't forget to subscribe to Android authority don't forget to hit that Bell icon so that you get notifications when we publish a video and please do go over to Android authority.com because we are your source for all things Android\n"