Shally's Alley

Hello everybody. All right, I think I have everything set up. Let me make sure I've got the video resolution good. Microphone, yes, check. All right, and output also check. Excellent. All right. Well, let me see if this is— social comments display here, click on them to showcase on screen. Okay. So I'm gonna see if this is actually live right now. Oh, Glenn says that it's live. Excellent. Okay, well, welcome everybody. This is my first time using our new platform. This platform is called Restream, for those of you that might be curious, and what's interesting and really cool about this is I can actually publish the feed to multiple channels, so this can go into the group It can also go on LinkedIn Live and on Twitter through Periscope as well as YouTube, so it's pretty much going everywhere, and it basically shows me how many people are live across all the different channels, so that is pretty cool. Unfortunately, I don't know if I have the ability to chat with everyone, so I will probably need to do private chat here or just, you know, hopefully go Layne can keep an eye out for any questions that are coming up via the chat until I figure out how to enable universal chat across all the channels. So here we go. This is the first ever Restream live Shelly's Alley, and today's topic is reverse image search. Reverse image search. Another one of the nice things about this is I can actually do this kind of picture-in-picture stuff here for those of you that still want to see me. I know it's not a big deal, but but here I am anyway. So the first question that it's probably going to come to mind is, what is reverse image search? Or rather, what does it mean and why is it important to me? Think of it as starting with an image instead of text. So it's the same concept as what you would normally do for search, except instead of using words or typing in words, you're going to provide an image. That may be a little bit confusing because you're probably wondering, well, what kind of an image are we going to use? Well, you could use all kinds of different images. For example, you could provide the images of a certification icon or some sort of application, software application, or product, a company logo, a product logo, profile photos, photos of buildings and locations. These are all pictures of things that you can provide. Why would we do this? Well, this is part of what we generally consider to be, in the big picture, to be part of deep web search. Although normally when I talk about deep web, I'm usually— I usually refer to content that is not accessible through your mainstream search engines like Google or Bing. What makes this deep web is that we're using, in some cases, other search engines not necessarily Google or Bing, although I'm going to show you how to do this with Google and Bing, but we're accessing content that is not identifiable through text search or through the main Google search interface. So even though we might use Google or Bing, we're using a different database of content. This has been discussed in, in various episodes previously as far as like why that content is actually different. So one of the first places I want to take you is to a website called Tineye. You might be familiar with Tineye. I'm going to type it here into the URL search bar here so you can see, and it'll be recorded on the video. But when we go there, I'm going to tell you a little bit about what this site does. This was my first experience using reverse image search way back when. I think— I don't even want to admit how long ago. And this remains still to, to this day one of my favorite tools. If you're a big fan of Chrome, you can actually add add this as an extension to where you can right-click an image and reverse image search it directly from inside of Chrome. So you'd be able to say right-click on this robot, although I'm using a Chrome browser that's got no extensions on it so that I don't have any slowness in the video. Otherwise I would show you, but if you added this to your, to your extensions, you would simply be able to right-click and reverse image. So now that we're here, there are two main ways to use TinEye, and most of the other tools that we'll talk about today. That is to either provide a photo, an image of any format. Well, let me take that back, not any format, most common formats, PNG, JPEG, GIF, those kinds of things. I'm sure there's a few that they probably don't cover. So you actually either have the image itself or you have an address, a universal resource locator or URL for that image. Whichever one you have, you can provide it either through uploading it or through entering the URL to the image. So let's start with an example for perhaps certification. One of my favorites is to look for the CCIE. So for those of you that might be familiar with CCIE, that is the— well, I'm not gonna be able to say that today, am I? The Cisco Certified Internetwork Expert, for the essentially Cisco engineer, Cisco networking engineer. One of the ways that you can find these logos are by going directly to perhaps the, you know, Cisco certification website, and they'll have a page that shows their, you know, media. It might be like a media page or download our logos or something like that, but you can even just basically do a pretty much short and sweet search on Google Images for basically CCIE logo or CCIE image or whatever, and you'll see that there's quite a few of them out there. I'm going to start with the network one. So there's, there's, there's a bunch of different types of certifications. This one here is the routing and switching one, so I'm going to copy the image address. And bring it over here. Brief, I'm gonna show you briefly where the image is. So this is where it's located. This is the home of the image. If I hit enter here, it's gonna basically just load the image itself right there, which is not necessarily what I wanna do. I want to provide that image here as a URL. So I pasted the URL here rather than up here in the address bar. So I'm not browsing to that image. You can see that it ends in.png. This tells us that it is in fact an image. And I'm going to hit the magnifying glass icon. This will then search through various databases to try to identify where this particular image is also being used. And you can see that it begins to provide a lot of different websites. Now, um, I kind of tricked you a little bit because this is a pretty common one, right? There's a lot of CCIE people out there, so it's giving me you know, 511 results out of 45 billion pages. But I wanted to kind of show you the variety and the depth that this goes to. So obviously you might want to narrow things down a little bit, maybe use a more unique logo or something like that. But the question remains, why are we doing this? Why can't we just search for CCIE? Well, if you search for CCIE, you're going to get all kinds of mentions, articles and things like that. But by searching for the actual logo, more than likely you're only going to find this logo on the personal pages or the blog pages or LinkedIn profiles or whatever of people that are using this as a badge and sharing that they are themselves CCIE certified or Cisco certified. Now the other side of this could be that they potentially may be a company that's enterprise certified, like a Cisco enterprise certification. So you fix that essentially by, you know, making sure that you've got the right logo. The other way to do this, which I'm gonna just go ahead and show you with the same example, is if you have already saved an image from somewhere, it could be from a Word document, it could be— basically the sky's the limit. I mean, anywhere that you can find an image, you can download it and save it, and then you can upload that image and use that image as the search criteria. So I'm going to click upload here. I'm actually going to use the same one because I'm just kind of showing you how this works. And I upload the image, probably going to have exact results. I don't see why there would be any difference. And you can see here it's doing the same kind of thinking through all of its, um, all of its sources. And there you have it. To me, the value in this is in identifying pages and websites that are well hidden. They're out there somewhere in the, whatever you want to call it, deep web, you know, dark corners of the internet, whatever. And it's giving me often hints of places like here's a course. Well, this is probably a course website where they teach CCIE. I would go there and I would look at perhaps the students, maybe they have a discussion group, and I might be able to find people that are studying for their CCIE certification. You might also find communities and, and blogs and other types of websites that you may not have even considered. Once again, you're, you're, you're really short-circuiting your search because instead of having to plow through lots of textual content, you're going right to the pages that have this exact image on them rather than are just, you you know, text talking about this certification. So that's an example for certification. Another one could be an application, like say for example Red Hat. Red Hat, you know, there's probably various different flavors. So let me see if I can find one here. Just going to use the actual Red Hat logo directly from them. And I'll show you, I'm going to download this, or essentially when you have an image from your, your computer, you can just move it into a folder or whatever. But when it's online and you see it, you can simply right-click and copy the, the address. So you can right-click and copy image address, or you can right-click and save image as. Whichever one you plan to use. For this case, I'm going to do the download because I want to show you an example of using the upload link. So I'm gonna upload this here and find my Red Hat picture and upload it. So this is rather than a certification, this is a an application. The logo of arguably it's a company, right? But it's a company that produces an application for the people that work with Linux. So here we get places like, well, here's Eclipse, which is the— oh, Glenn says you're not seeing it. Do you see the Red Hat now, Glenn? Okay, okay, great. There might be a little bit of a lag, I think, with the, with the stream, so note to self, slow down a little bit, wait for the images to load before I start talking. So anyway, here is a Twin Cities CIO magazine, probably, and it looks like if I browse there, this looks like it's a page that shows who sponsors this Twin Cities CIO group. So that's a place I might go to find people that are Red Hat users or etc. So here's what comes up when I, when I go to that page here. Of course, this is a sponsorship page, but think about what else I can see here. I can see events, I can see awards, and so on. So the, you know, if you're looking for CC— for not CCA, I'm sorry, if you're looking for Red Hat Linux people, this might be a really good source. Another type of, another category of image could be, for example, a product logo. You know, you might be looking for people that are product managers, or you might be looking for people that have familiarity with a certain solution set or product, company logos as well, similar to what I did with Red Hat here. And another use case for me that I frequently have to, you know, look for is people's headshots. Now, why would I need to use someone's headshot? Well, often I do sourcing on Twitter and I find that the Twitter handle or the Twitter username isn't really very revealing. It doesn't tell you who this person is. It may be a nickname or it may even be perhaps the name of the company or they might have a small business that they use, you know, on the side for consulting. But bottom line is the headshot is the headshot of that person. However, the account name doesn't really tell me very much. So if I were to look for people on Twitter that are, let's say, you know, Red Hat certified, do a quick search on that. Let's see a bunch of Red Hat services here. Unfortunately, one of the downsides I'm finding here is that when I have one tab open that's streaming, I can't really stream the other tab, but that's okay. I'm going to grab something from here real quick and bring it over. I'm just doing a search for Red Hat Certified. On Twitter and gonna pull up a profile of someone that doesn't have a, you know, their name, for example. Okay, so here's an interesting one. This gentleman's profile is Oz Efendi, which is very likely a nickname or a personal handle that they're using. As you can see here, they're a Red Hat Linux admin and also are familiar with Oracle DBA and VMware, but it doesn't tell me what their name is. However, I can use the image as a search to see if I can find perhaps their Facebook, their LinkedIn, or blog or something else. Now sometimes you might see right here, he's already got a blog, so let's pretend that we didn't have the advantage of this. Obviously, the first thing I would do would be to go to this blog, or if you had a LinkedIn page linked here, I would do that, but let's pretend that that doesn't exist and all we have is this image. So once again, you can either copy the image address or save the image. If we copy the image address, then we simply go back to TinEye and paste that image here. Looks like it didn't do it. There we go. Alright, so now I found this on two places, Foursquare and Kiva. Kiva is a nonprofit organization, donation organization, so I'm going to go check that out. And Foursquare, of course, as you might know, is a way that people check in, is probably now Swarm, but the website is still Foursquare. So I might be able to identify their full name or maybe even contact information by seeing their Foursquare profile. So those are two places that I found this particular individual. Unfortunately, I just looked at the, um, Kiva and it is anonymous. So he's got his picture on there, but it's still anonymous. And I went to the Foursquare and it doesn't have his name. But you can see how this would be very helpful. In this case, we already have the blog, so we didn't really need it. But in the case where you have nothing, you might be able to pull up you know, something else basically, even another Twitter account, or it could be a headshot on some, you know, webpage, you name it. There's all kinds of stuff, interesting stuff that you can find out there that'll be really revealing. The last use case I want to discuss is to use buildings or locations, building photos or location photos. This is useful when you're trying to identify, once again, you have the an image of a building, but you don't know where it is, or you don't know what it is, and it's related to, you know, someone. Maybe they've posted a photo on any one of the photo sharing websites, and they're standing in front of a building, and it says, you know, here's where I work, or whatever. I don't know, I mean, it could be— it could just be a casual photo of them right in front of their building. You have no idea what building that is. You can just use the building, the image of that building, as a reverse image search and find out what that is. Might even tell you the name of the company. So for that, you need the picture of— it doesn't matter if the person's in the picture, you can just crop out the piece that has the, the important content, right? So it'd be like a, like, um, um, like an office building, for example. I'm just going to do a quick search for office building and find just a generic one. I have no idea what this is. I literally just put in office building, and I'm copying the image of the very first building that came up. So literally, if I could show you this, you would see I just saved the very first image. I have no idea what this is, and I'm gonna go over here and just upload that image that I just downloaded. By the way, you can also grab the URL. I'm really Just depends on how, how much you're going to be doing with this. If you're going to use other sources, you might want to download a copy of it. So here we see that image I knew nothing about, and now it's showing me a couple of other places where this image appeared. I'm finding out that this is now a building in, in the Netherlands, so that's a piece of information that I did not have before. I can browse to both of these Pages, and unfortunately I'd have to grab the addresses here and bring them over and see what I can find. So one of them leads me to an architecture.net magazine that has, as you would imagine, this building featured in one of the news articles, and it tells me basically where it is. So now I know where this building is and what this building is. Imagine if there was somebody that was blogging about it or talking about it on their posting about on their Twitter or whatever. And you didn't know what it was, now you have that. So you may even be able to certify from there who their employer is, etc. This is the one that I saw when I opened up and it didn't— see right there? Land handling technologies in the Netherlands. So I could— when I opened the other page, I couldn't bring it over, but this is what I'm seeing here. Land handling technologies in Burki and Schott, Netherlands. So those are the different use cases. Now, is there something besides TinEye? Yes, TinEye has its own algorithm. It's one of the oldest ones as far as I know, certainly my go-to, but it's not the only one that does this kind of search. There's also another one called Bearify, that's with a B, and I'll take you there real quick. Similar to TinEye here, you can either upload an image or you can paste the URL, and this uses a different database, a different set. Let's actually upload the, the building that I just uploaded, and we know what the two websites were that came up before, so let's see if it's the same two, or again, I've never done this before, so let's see. Ah, it's asking me to log in, so this is my clean browser that doesn't have any, any memory in it at all, so I'm gonna have to re-register here, and I don't want to take the time to do that, but basically that would be an example of something that you could try on your own. I don't want to take time to go and do that because I have 4 other places, actually 3 other places I want to show you. Next one is of course Google, Google's own image search. When you go to Google Images, or you can either go to images.google.com, or I believe you can also go to google.com/images, you'll see a little camera icon. By the way, if you have the app on your phone as well, you can actually use the app on your phone for the very, very same purpose. What it would do is it would actually allow you to either, you know, upload from your media or take a picture. So I don't know if I can, let me see if I can do this. This is kind of low-tech, but it might work, let's see. Just on a lark to see if it actually works. Okay, I'm going to minimize this and maximize that, okay, see if I can do this. There it is, so you see, you see, this is so weird, you see right there, well, you can't see right there, 'cause I just touched something, never mind. There, okay, where it says search, That's the, that's the app, and you can just click on the lens, the one that, that says lens over there, and that will take a picture of whatever you're looking at in the real world, or, you know, it could be a picture of some of the document or whatever, and it'll search for that. Same thing can be done here with Google Images by clicking on the little camera icon, and once again, you have both choices. You can either paste the URL, or you can upload the image. Let's go ahead and grab that building and see what we can find that may be different than Tenai. Remember, I've never done this before with a particular image. I've never used this image before, so I have no idea what to expect. Flying by the seat of my pants here. Okay, it tells me it's an office building. Well, if it wasn't for Google, I wouldn't have known. So what— but, you know, of course, all kidding aside, what I wanted to show you is that down below here you'll see other pictures that are— some of these are similar, but we could actually match this exact same thing by looking through these. Now I'm not seeing one that's an exact match, but there's one that says pages that include matching image handling technologies. Look at that. There it is. This is actually the building. Tells me who the architect is, not that I needed to know that, but it actually gives me— oh, I've even got the inside. Tells me a little bit about what the building is being used for. There's the floor plan. I mean, this is excellent. So that's an example of using buildings and locations. The same thing happens for like, think of a landscape, right? You've got a cityscape behind you. And— or behind, I should say, the person that took the picture, and you have no idea where that is. You want to know what city that is by the skyline. Well, if you take the skyline of that city, it would— it would be able to match other skylines that are similar. So for example, let me grab this one right here. Taking me a second to download it. I should have probably picked a smaller file. And I'm just calling it skyline. I still don't know what this is. I just did a search for skyline and that's it. And now I'm going to go back to Google Images and click on the camera. Upload an image, and choose my skyline. Let's see where this is. It says cityscape, okay. Visually similar images, oh, that's a little A little hard to tell, some of these look pretty similar, but let's see what it says. Matching images down here. Okay, look at that, Singapore. Boom. You know exactly where it is now. How cool is that? It doesn't have to be, it could be a park or whatever, and you know, you'll be able to identify where that, where that is. The other search engine that does this very well that I want to share is Bing. So Bing has a different database than Google. It operates differently, um, and has, you know, different algorithms altogether. So you might get some of the same results, or you might get different results. With Bing, you click on the image icon over here that's next to the magnifying glass. So you click on the search using an image. Once again, you can paste the URL, or you can upload your, your photo. You can drag a photo and upload it here, just drag and drop, or you can click browse Let me see if I can, actually, I'm just gonna click browse. And apparently here you can also take a photo, so I'm assuming that they have an app for that as well, and I'm gonna upload the skyline. Look at that, off the bat, it immediately told me, oh, this is Singapore cityscape. So Bing is another one, you know, just a different set of data, you might find an icon or a person's headshot on one but not another. Let's see, let's upload the picture, excuse me, the picture of the, of the individual, oh, I don't think I downloaded that, did I? Oh, that's okay, never mind. I didn't save the picture of that, of that person from Twitter, but I was going to try and upload that and see what we get. All right, before I, you know, take up any more of your time, let me go to one other place, and that is Pinterest. That's right, Pinterest. You might not think of this as an image search, but they actually also have a reverse— I don't know if that's the right way to think about it on On Pinterest, but they have an image search using images. Come on, bring me down to more of my images. Hey. Okay, fine, I'll just do a search. Let's do CCIE. All right, I'm gonna open this up. No, that's not a CCA. That's weird. Okay. Let me add Cisco to that. There we go. Okay. So, if I open up the image, this little icon down here with the sort of rounded corners and the micro— the magnifying glass. If you click on that, it essentially does a reverse image search. So what it's doing is it's finding more ideas, but basically it's using the visual elements and trying to find others like it. And you can actually crop a piece of the image right from here. So this might even be a quicker way to do something like, uh, you know, like a city or something, because you can grab the corners Let me see if I can find something a little bit more useful than that. Let's try Cisco, perhaps. Cisco networking. I'm trying to find like a, you know, like a stack of servers or a stack of routers. Ah, here we go. Cool. That works. Let me try this. Okay. I'm going to use the image search, and what I'm going to do is I'm going to cut or sort of hone in on the part of the frame that I'm interested in, which is this particular, you know, the actual gear, the equipment, and see what it's doing is it's bringing back all these others, and these might very well lead to people's profiles. So you can use this for sourcing original leads or for just tracking down people, you know, etc., different uses like that. So to recap, we're about a minute past our half hour here. Just a quick recap, we covered different ways that we can use images to source by looking up people's photos, for example, by looking up certification icons or certification images, graphics, the application itself, like Red Hat, or the products. It could be, you know, a hard product, a software product, product logos, company logos, people's profile headshots, people's profile photos, buildings, and other locations. And we covered how to do this with my favorite, of course, tineye.com, which allows you to right-click and reverse search an image, but also with Berify, b-e-r-i-f-y.com. And I wasn't able to show you that because it required me to log in, um, and we didn't have time for that, but also images on Google and images on Bing and the icon at the bottom right corner of an image on Pinterest. If you have any other ideas, drop me a line. Would love to hear about your experiences with image search. Until then, thank you very much for attending our live session today, and leave some comments. Let me know how it went. This is my very first time using Restream. By the way, if you want to get the handouts, the notes that Glen takes, and the attachments and everything else, you have to join the Facebook group. Even though this is streaming everywhere, um, I'm only going to upload them to one place. So if you want the notes, go to the Sourcing Method on Facebook, join. It's absolutely free to join, and you can go in the file section and you'll be able to see a recording of this as well as access the notes and any other attachments that we might have, like PowerPoints, etc. All right everybody, thank you very much.