Thursday, March 30, 2023
Bringing the Latest in News Straight to Your Screen

Using AI Self-Driving Cars To Detect And Help Curtail Eyesore Graffiti Around Us

By News Creatives Authors , in Business , at August 15, 2021

Graffiti can invoke quite heated debates.

Some assert that graffiti is a blight, a blemish, altogether offensive, and ought to be condemned as a practice. Many municipalities have laws on the books that make it illegal to produce graffiti. It is considered a form of vandalism. Besides creating an eyesore, there are lots of other qualms about allowing graffiti to exist and potentially flourish.

For example, arguments are made that graffiti is oftentimes a precursor to worse crimes. If you allow graffiti to be marked here and there, you are presumably saying that lawlessness is being tolerated and potentially even encouraged. This is the contention behind the slippery slope theories about graffiti. Once you start falling down the graffiti abyss, it can end up badly and generate a soul-crushing crime spree of shocking proportions.

A contrarian view is that graffiti is art. If you try to censor or prevent graffiti, this is tantamount to suppressing freedom of expression.

Furthermore, the argument sometimes goes, the display of graffiti is nothing more than letting off a bit of harmless steam. When you try to keep the steam bottled up, it will merely boil and eventually explode, perhaps transforming into wanton crime. Thus, the claim is that attempts to stop graffiti are what leads to crime, rather than the acceptance of graffiti upon our public facades.


One thing to keep in mind is that graffiti is apparently about as old as the capability of humanity to mark on walls, perhaps including primitive cave walls. There is certainly substantial evidence that graffiti seemed to exist in ancient Greece and during the Roman Empire. Of course, just because graffiti has been around for a long time does not necessarily make it right.

Again, you can posit postures both for and against graffiti.

Law enforcement tends to point out that graffiti can be a type of territorial marking by street gangs (frequently accompanied by tagging or including specific markings connoting the particular gang). In that case, the notion is that the otherwise “innocuous” graffiti has shifted out of the artsy framework and become a crime-inducing tool. A street gang that marks with graffiti is likely to be willing to defend their turf by armed means. A rival gang is bound to take offense at markings that are displayed in their territory. The inevitable clashes can produce deaths and injuries, not only among the gang members but extend into the innocent public at large.

On the other hand, in some stifling political settings, the use of graffiti might be the only viable means for a suppressed populous to express their beliefs in any publicly seen way. The spray painting or marking that is undertaken to produce graffiti is usually done in a secretive manner, doing so at nighttime in darkness or when others are asleep or not watching. This provides a semblance of anonymity, which could be vital if there are powerful political forces that would harshly stifle those wishing to voice alternative beliefs.

Back and forth we can go on the merits and demerits of graffiti.

Where does graffiti usually appear?

I’m sure that you’ve seen graffiti in a wide variety of places. We almost seem to shrug our shoulders somewhat and accept the idea that there will be graffiti on the walls of subways or inside pubic bathrooms. That just seems to happen as though by some mystical force.

When driving your car, I’d bet that you come across graffiti if you are driving with any regularity and especially when driving in busy cities or anyplace of a sizeable population. There is graffiti written or splayed onto the walls of office buildings. There is graffiti on bus benches. There is graffiti at times on traffic signs such as Stop signs that are marked with graffiti.

You might think that being on a freeway would somehow avoid any chance of seeing graffiti. Not so. There is likely graffiti on the walls that serve to guard the freeway and protect from pedestrians or others meandering out onto an active byway. There is graffiti on the overpasses. There is graffiti on the underside of overpasses. And so on.

In fact, while driving, you are bound to see graffiti in every which way. Upon driving along, you might see graffiti on the walls of the freeway. In addition, you can clearly see buildings that are built adjacent to the freeway, and perhaps the walls that front the roadway are coated with graffiti. The traffic directional signs that tell you what the next exit will be can have graffiti on them. Etc.

You almost have no choice but to see such graffiti.

Why so?

As a driver, you are obligated to keep your eyes on the road. This includes the entirety of the driving scene. Even though you might be concentrating on the cars and trucks ahead of you, nonetheless you are cognizant of the other roadway objects too. Your eyes are gazing all around, looking to make sure that there aren’t any surprises that might suddenly alter your driving efforts.

I’ve known drivers that mentally blotted out the graffiti after seeing it day after day. In other words, when driving in a new area that you’ve not previously encountered, you are likely to be hyper-aware and be looking at everything around you. Once you’ve made the same commute dozens of times, you begin to only notice something new and unusual. If there has been graffiti there the entire time, this tends to go into your mind as not pertinent to the driving task and it mentally gets disregarded.

That doesn’t mean that out of mind also means out of sight.

You still can see the graffiti.

If someone were riding with you and they had never gone that route, they are bound to point out to you that there are numerous graffiti markings. At that juncture, you might be jarred out of your complacency and suddenly see anew that which had begun totally familiar and humdrum. The other person might comment on the question of how you can tolerate all that graffiti, and the verbalization of the qualm is enough to knock you out of the fog that blanketed the graffiti in your noggin.

There is a whole lot more to the matter of graffiti and we ought to cover herein some additional insights worth noodling on.

Some stridently point out that graffiti can produce negative environmental effects. There is the visual blight involved, assuming that the graffiti is seen as an eyesore. The graffiti might be produced via the use of toxic chemicals and therefore produce adverse environmental consequences. Consider too the effort to later remove the graffiti which can also consume and generate toxic mixtures and cause disposal nightmares.

A local municipality might not be able to afford the cost to get rid of graffiti in the sense of bearing the burdensome costs to find and remove the graffiti.  Someone making graffiti can do so relatively cheaply. The labor and effort to remove the graffiti can be very costly. There is a decided cat and mouse game that often occurs in the graffiti-producing and graffiti-removing gambits. The problem for those removing the graffiti is that the price tag to do so is woefully out of proportion to those that are generating the graffiti (cheap to make, costly to excise).

To try and cope with graffiti, some administrations will create a database that catalogs the graffiti in their township. This provides a means to track where the graffiti is appearing, along with noting when it will be removed and tracing the costs and labor consumed to do so. Patterns can be potentially uncovered by analyzing the database info.

For example, perhaps a database analysis reveals that graffiti in a particular city tends to be applied to buildings that abut the freeway and for which there is no fence around the building and little or no lighting at nighttime. If that is the case, there is a chance that some modest preventative efforts might aid in curtailing the graffiti. By getting the building owners to put up fences and add lighting, it might shut down those surfaces from being so readily exploited by the graffiti makers.

Lamentedly, in the ongoing cat and mouse ploy, the graffiti makers might perceive that those erected blockages are merely a challenge to be merrily overcome. As such, they might cut through the fences and attempt to knock out the lighting.

Or the graffiti makers might decide to find places that are less difficult to mark, and ergo opt to do their graffiti elsewhere and you’ve not reduced the graffiti impulse per se. Some pundits worry that various “defensive” efforts to curtail graffiti that only focus on ways to make graffiti harder to apply simply push the graffiti makers around town, spurring those makers into finding new spots. Thus, they are not being discouraged from continuing to proffer their graffiti. It is for that reason that some argue you need to get at the root causes that are stoking the graffiti, and presumably, rein in the graffiti by lessening those factors.

The databases that are used to track graffiti can be utilized in a multitude of ways.

Police might analyze the data to gauge where various gangs are setting up the turf. In addition, when the markings of one gang start appearing into the presumed sphere of another gang, the police might be able to anticipate that gang-related shootings and stabbings are going to arise. This can give a heads-up for those community programs that are trying to solve local gang confrontations. Plus, law enforcement can beef up efforts to try and stop the skirmishes before they erupt.

When you are driving your car in an area that has graffiti, it can be surprising to see the graffiti and act as a wake-up call about where you are. In some instances, this might be a safety-oriented wake-up call. In other cases, perhaps the wake-up is that there needs to be more done to cope with the underlying causes that are spurring the graffiti efforts. One supposes that at times, the graffiti can also be seen as a work of art and admired for what it portrays.

There is one thing about graffiti that we can nearly all definitively agree with.

For drivers, graffiti is a potential distractor from the driving task.

Despite whatever else you might believe about graffiti and whether it is a rightful practice or a wrongful practice, it can be a distractor to drivers. A person at the wheel of a car should be paying attention to the roadway and all of the objects in the driving scene that is directly pertinent to the driving chore at hand.

The graffiti is not adding value as an indicator of the roadway status. You could argue that it is doing much worse. If the graffiti is marked on traffic signs, this might make it harder to discern what the traffic sign is meant to convey. Any delay in being able to properly read a traffic sign can produce driver errors. Driver errors can lead to getting into fender benders or all-out car crashes.

On top of this, a driver that tries to look upward at graffiti painted on the underside of an overpass is tempting fate in a very terrible way. While looking up at the graffiti, the car or truck immediately ahead of them might be applying their brakes and this distracted driver is unaware of that impending action. Wham and the distracted driver smashes into those other vehicles ahead.

I’m sure that some smarmy retorts would be that only drivers that are dolts would be distracted by graffiti. The argument would be that any driver that is worth their salt ought to be able to drive fully and unencumbered despite whatever graffiti might exist around them.

The problem with that retort is that anyone making such a claim does not seem to be driving in the same real-world as the rest of us. Drivers are distracted by all kinds of things and you cannot make a blanket claim that only bad drivers do so. Added to this distraction is that even a careful driver might be fooled into thinking that the graffiti is some form of roadway indication, or in the case of marking on roadway signs the graffiti is purposefully preventing the driver from being fully attentive.

My point is that you would be hard-pressed to contend that graffiti is not distracting to drivers. And you would be further hard-pressed to argue that drivers should miraculously ignore the graffiti as though it isn’t there at all. Pardon my French, but face up to reality, please.

Graffiti that is visible to drivers is adding risk to the driving task.

You can try to debate what level of risk is being derived, and you can assuredly point out that there are lots of other distractors in the everyday driving arena, but the bottom line is that graffiti is a distractor. Whatever other value you might place on graffiti, it has to also accept the unfavorable burden of being a driving distractor. That is rather cut and dried, I’d say.

Speaking of cars, the future of cars consists of AI-based true self-driving cars.

There isn’t a human driver involved in a true self-driving car. Keep in mind that true self-driving cars are driven via an AI driving system. There isn’t a need for a human driver at the wheel, and nor is there a provision for a human to drive the vehicle. For my extensive and ongoing coverage of Autonomous Vehicles (AVs) and especially self-driving cars, see the link here.

Here’s an intriguing question that is worth pondering: How will AI-based true self-driving cars handle graffiti that is visible in the roadway scene, and how might we leverage the advent of self-driving cars toward society contending with graffiti?

Before jumping into the details, I’d like to further clarify what is meant when I refer to true self-driving cars.

Understanding The Levels Of Self-Driving Cars

As a clarification, true self-driving cars are ones that the AI drives the car entirely on its own and there isn’t any human assistance during the driving task.

These driverless vehicles are considered Level 4 and Level 5 (see my explanation at this link here), while a car that requires a human driver to co-share the driving effort is usually considered at Level 2 or Level 3. The cars that co-share the driving task are described as being semi-autonomous, and typically contain a variety of automated add-on’s that are referred to as ADAS (Advanced Driver-Assistance Systems).

There is not yet a true self-driving car at Level 5, which we don’t yet even know if this will be possible to achieve, and nor how long it will take to get there.

Meanwhile, the Level 4 efforts are gradually trying to get some traction by undergoing very narrow and selective public roadway trials, though there is controversy over whether this testing should be allowed per se (we are all life-or-death guinea pigs in an experiment taking place on our highways and byways, some contend, see my coverage at this link here).

Since semi-autonomous cars require a human driver, the adoption of those types of cars won’t be markedly different than driving conventional vehicles, so there’s not much new per se to cover about them on this topic (though, as you’ll see in a moment, the points next made are generally applicable).

For semi-autonomous cars, it is important that the public needs to be forewarned about a disturbing aspect that’s been arising lately, namely that despite those human drivers that keep posting videos of themselves falling asleep at the wheel of a Level 2 or Level 3 car, we all need to avoid being misled into believing that the driver can take away their attention from the driving task while driving a semi-autonomous car.

You are the responsible party for the driving actions of the vehicle, regardless of how much automation might be tossed into a Level 2 or Level 3.

Self-Driving Cars And Graffiti Around Us All

For Level 4 and Level 5 true self-driving vehicles, there won’t be a human driver involved in the driving task.

All occupants will be passengers.

The AI is doing the driving.

One aspect to immediately discuss entails the fact that the AI involved in today’s AI driving systems is not sentient. In other words, the AI is altogether a collective of computer-based programming and algorithms, and most assuredly not able to reason in the same manner that humans can.

Why this added emphasis about the AI not being sentient?

Because I want to underscore that when discussing the role of the AI driving system, I am not ascribing human qualities to the AI. Please be aware that there is an ongoing and dangerous tendency these days to anthropomorphize AI. In essence, people are assigning human-like sentience to today’s AI, despite the undeniable and inarguable fact that no such AI exists as yet.

With that clarification, you can envision that the AI driving system won’t natively somehow “know” about the facets of driving. Driving and all that it entails will need to be programmed as part of the hardware and software of the self-driving car.

Let’s dive into the myriad of aspects that come to play on this topic.

The first aspect to cover encompasses the detection of graffiti by self-driving cars. If self-driving cars aren’t able to detect graffiti then the case is summarily closed and we can move on. Thus, detection is vital.

Self-driving cars will be outfitted with a sensor suite that is used to detect the driving scene. Devices include video cameras, radar, LIDAR, ultrasonic units, and the like. These sensors are active while the self-driving car is underway and feed data into the onboard AI driving system. The AI driving them uses the data as though the sensors are essentially acting as the eyes and ears of the autonomous vehicle.

Many people seem to falsely think that self-driving cars will only be sensing whatever is immediately necessary for purposes of driving the vehicle.

That kind of narrow sensing is not especially in the cards, as it were.

In reality, the sensors are capturing a wide array of data that is far beyond just sensing a driving scene. When a self-driving car goes down your neighborhood block, keep in mind that the sensors are likely collecting video that showcases the children playing in front yards and people tending to their own efforts while meandering outside their homes. I have been referring to this vastness of data collecting as the roving eye, see my discussion at this link here.

The data being collected by a self-driving car is customarily used to guide the driving actions of the autonomous vehicle. That is what the stated purpose of the sensors seems to be. Nonetheless, the data collected can be used for a myriad of other purposes. I’ve emphasized that this data can be monetized by the fleet operators of self-driving cars and that we have yet to firmly establish who owns this data.

Undoubtedly there will be privacy intrusion claims raised (see my analysis at this link here).

The good news about the data collection of self-driving cars is that they can potentially be used for crimefighting purposes. Imagine that the data from zillions of roaming self-driving cars gets uploaded into cloud databases. Those cloud databases are then assessed to ascertain what happened at crime scenes, in the same way, that law enforcement today might try to find nearby security videos that perchance captured something of relevance. Once we have self-driving cars aplenty, it is going to be a massive and readily used means of figuring out where a criminal committed a crime, along with where they went afterward. See my discussion at this link here.

What does this have to do with the detection of graffiti?

That’s easily answered.

You could piece together the video camera recorded imagery from self-driving cars and do computational pattern matching to try and identify graffiti. All those self-driving cars driving down an everyday freeway would have a video showing the freeway walls, the buildings adjoining the freeway, the overpasses, and so on. Algorithms could be readily applied to the collected video data and indicate where graffiti is visibly present.

The difficult part of this is figuring out what constitutes graffiti.

A billboard that has a funky-looking display could be misconstrued as graffiti. Graffiti that is put on top of a billboard that has some other advertising imagery might blend in and not be easily asserted as graffiti. In short, graffiti can at times be ambiguous, even when completely visible to the naked eye.

On the other hand, humans pretty much can discern graffiti when they see it. In that semblance, we could anticipate that crafting computer-based algorithms to detect graffiti will be somewhat straightforward to do. The detection would most likely include a set of probabilities to denote the likelihood that images labeled as graffiti were in fact graffiti.

You can expect that there will be some amount of false assignments. Something that appears to be graffiti might get marked with a high probability of being graffiti, even though it turns out to not be graffiti. That’s a false positive. There is also a chance that something not rated as graffiti is indeed graffiti. That’s a false negative.

On the whole, though there will be those mismatches or misinterpretations by the algorithms, you can expect that those will gradually become less so once the Machine Learning (ML) and Deep Learning (DL) approaches further improve. In addition, humans would likely be used to aid in guiding the algorithms and be able to catch the false positives and false negatives.

Let’s go ahead and agree that detection of graffiti by self-driving cars will be relatively possible.

Now what?

Well, similar to today’s simpler databases, graffiti detection could be maintained in an online database. Those that are seeking to cope with graffiti could use the database to explore where the graffiti is being placed. Crimefighting such as the earlier depiction of police trying to identify gang activity could take place.

One perhaps not so obvious advantage of self-driving cars being used to detect graffiti is that it is almost a no-cost added-value proposition.

Self-driving cars are already going to be collecting the visual video scenery, doing so for purposes of driving the autonomous vehicle. The data can be stored locally onboard the self-driving car and then uploaded when it is possible to undertake an OTA (Over-the-Air) electronic communication. This is pretty much going to be happening anyway.

From an added cost perspective, the main aspect will be devising algorithms to detect the graffiti in the visual imagery captured and uploaded. This won’t be particularly costly. There is the cost of storing the data, but that’s also relatively straightforward in today’s large-scale cloud environments.

Presumably, the fleet operators could charge fees for those that want to access the data that has been collected from their self-driving cars. A city that wants to analyze the data for graffiti would pay a fee or possibly subscribe to the database. Same for a law enforcement agency. There might be third-party companies that formulate such systems to provide this capability for those that want to tap into a graffiti tracking online system.

Perhaps nonprofits would get a special break on pricing for making use of the graffiti databases, assuming that the entity was striving to help reduce graffiti and was working with the local community in doing so.


Recall that earlier there was an important point raised that graffiti can be distracting to human drivers.

You cannot generally say the same for AI driving systems. In short, the computational processing of the driving scene is already trying to pinpoint what is important and what is not important when seeking to drive a self-driving car. The graffiti is not especially distracting, unlike for a human that might linger mentally on the graffiti and ponder all sorts of deep thoughts about the significance and meaning of the graffiti.

Let’s be clear, I am not giving a free pass to graffiti. The AI driving system and the onboard hardware processors are going to have to ferret out what the flaunted graffiti is perhaps hiding or what it might portend related to the driving task. There is that kind of cost involved. Nonetheless, it isn’t anything on par with the type of distraction that would be exhibited by human drivers.

If it is the case that self-driving cars are relatively unaffected by graffiti, maybe this would spark those graffiti makers into becoming wilder and bolder, some might worry.

Maybe so, maybe not.

Via the use of automation to track graffiti, the odds are that various other societal forces would have a much easier time trying to find ways to discourage or excise the graffiti. The cat and mouse game slightly begins to tilt towards those cats trying to stop or catch the mice.

And there’s something else to realize too.

Self-driving cars might end up being built as though they are bubbles. You get into a self-driving car and it has no discernable windows. There isn’t a need for windshields and nor windows, since there isn’t a human driver at the wheel. Passengers might prefer the privacy of being inside a bubble. If they want to look outside, all they need to do is display whatever the video cameras are seeing, and this can be shown onto the interior LED displays.

This subtle but poignant point is significant to the nature of graffiti. Will makers of graffiti make their graffiti if no one is looking at it? This is almost akin to the infamous question about a tree falling in the forest and whether it will be heard if no one is around (I’ve discussed this matter and answered this in my columns).

Pedestrians can still see the graffiti. But anyone riding inside an autonomous vehicle on the freeways or highways might not be watching it. Without that audience, presumably, a sizable chunk of graffiti would no longer be created. Or, as discussed earlier, the marking of graffiti would gravitate to places that primarily cater to foot traffic and thus where pedestrians would see it, rather than where roadway vehicles necessarily are principally found.

I realize that some might try to argue somewhat to the contrary on this emphasis about graffiti perhaps not being particularly paid attention to by humans inside self-driving cars.

Here’s the logic.

Since people won’t need to drive anymore, they can leisurely look at the passing scenery while idly riding as a passenger in self-driving cars. This might mean that there are eyeballs that today would be so preoccupied with the driving scene that they would give short shrift to looking at graffiti. But now that those eyeballs are unshackled from driving, they can gape in amazement at graffiti.

If that were to be the case, the graffiti makers would see this as an opportunity to make their marks known. The audience for graffiti will gradually become massive and attentive due to the emergence of self-driving vehicles. Time to get out the spray paint and start doodling on the walls and surroundings of our cities and towns.


As I said, we don’t know which way things are going to bend. You’ll have to decide for yourself whether the advent of self-driving cars is going to be principally a graffiti inhibitor or a graffiti promoter. In my book, tensions would seem to sway toward the graffiti disruption side of the equation. Time will tell.

Anyway, a future conversation about graffiti might go like this.

Hey, did you see that graffiti on the south side bridge?

The answer: No I didn’t, but I can check in with my self-driving car and find out what that’s all about.

Case closed.


Leave a Reply

Your email address will not be published. Required fields are marked *