Skip to main content

Mathematical Illusions: Ambiguous Objects

A math major is an excellent pursuit for any individuals time in college. Learning about the nearly uncountable concepts and ideas that are available to learn and master is staggering. This alone makes it a worthwhile investment. An added bonus of becoming a math major: becoming a magician. From card tricks to optical illusions, becoming a math major/magician can leave your friends and family speechless.

Ambiguous Objects

For this adventure, we will be looking at the optical illusion created by ambiguous objects. An ambiguous object is an object that looks dramatically different in shape, based on the angle at which the object is viewed. From one side of an object, a person may see a star, and from the other side, a circle. The following image shows just that:
Credit to Dr. Kokichi Sugihara to the creation of both the object and the image. This contrast in shape seems impossible to create from just a change in perspective; it almost seems like magic (or photoshop). This magic depicted is the math behind ambiguous objects in action.

The Math Behind the Magic

The first concept that is relevant to the trick is perspective. As cool as it would be, an object cannot morph into two completely different shapes. The object is one shape that roughly looks like some amalgam of the two shapes. This combination of the two shapes is created by placing two curves flat on a plane, and then angling them in such a way that takes advantage of perspective. For simplicity, it is assumed that the two angles that the object is being viewed is 45 degrees on one side, and 45 on the other. It is also assumed that the onlooker is far enough away that the view of all points of the object is 45 degrees. The goal is to have one viewer see one curve while the other viewer sees the other curve. For this example consider the following two curves:
The curves currently lie flat on the xy-plane, but a parameterization can be written such that one viewer sees a point on the curve denoted as (x, f(x), 0) and the other viewer, at the same point, sees (t, f(t),0). The line that is parallel to the 45 degree angle vector (0,1,1) and goes through the first curve can be written as \[h(s,t) = (t, s + f(t), s)\]. The other curve that is at the other 45 degree angle can be written as \[k(s,t) = (t, -s+g(t),s) \] where f(t) and g(t) are the curves flat on the plane. This definition leads to the parameterization of the ambiguous object: \[ c(t) = (t, .5(f(t) + g(t)), .5(g(t) - f(t))) \] This curve gives the necessary curve to create the object that looks like two different shapes and concludes the magical explanation.

Explaining the Example

To keep with the trend of magic, the example consist of a curve of a diamond and the curve of a spade. These two curves are different enough to show the illusion in action, and it came out quite nicely. The following two images show the opposite perspectives of the object in OpenSCAD:

The Curves

These curves were both created in a piecewise fashion. There are 2 parts to each curve for a total of four parts. The diamond is formed by the curves: \[ f_1(x) = \sin(x)^4 and f_2(x) = -\sin(x)^4\] Mapped over domains [0,180] and (180,360] respectively. The top part of the spade was formed by the curve: \[g_1(x) = .5|\sin(x)|^.5 \] With domain [0,180] and the bottom was formed over the remaining domain by the curve: \[g_2(x) = -\sin(x) \]

Comments

Post a Comment

Popular posts from this blog

The Approximation of a Solid of Revolution

Most math teachers I've had have been able to break down Calculus into two very broad categories: derivatives and integrals. What is truly amazing, is how much you can do with these two tools. By using integration, it is possible to approximate the shape of a 2-D function that is rotated around an axis. This solid created from the rotation is known as a solid of revolution. To explain this concept, we will take a look at the region bounded by the two functions: \[ f(x) = 2^{.25x} - 1 \] and \[ g(x) = e^{.25x} - 1 \] bounded at the line y = 1. This region is meant to represent a cross section of a small bowl. While it may not perfectly represent this practical object, the approximation will be quite textured, and will provide insight into how the process works. The region bounded by the two functions can be rotated around the y-axis to create a fully solid object. This is easy enough to talk about, but what exactly does this new solid look like? Is...
8 comments
Read more

Solids of Revolution Revisited

Introduction In my previous blog post on solids of revolution, we looked at the object formed by rotating the area between \(f(x) = -\frac{1}{9}x^2+\frac{3}{4} \) and \( g(x)=\frac{1}{2}-\frac{1}{2}e^{-x} \) around the \( x \) axis and bounded by \( x = 0 \) and \( x = 1.5 \). When this solid is approximated using 10 washers, the resulting object looks like this: When I was looking back over the 3D prints I’d created for this course, I noticed that the print for this example was the least interesting of the bunch. Looking at the print now, I feel like the shape is rather uninteresting. The curve I chose has such a gradual slope that each of the washers are fairly similar in size and causes the overall shape to just look like a cylinder. Since calculating the changes in the radiuses of the washers is a big part of the washer method, I don’t think this slowly decreasing curve was the best choice to illustrate the concept. The reason I had done this o...
Post a Comment
Read more

Finding the Center of Mass of a Toy Boat

Consider two people who visit the gym a substantial amount. One is a girl who loves to lift weights and bench press as much as she possibly can. The other is a guy who focuses much more on his legs, trying to break the world record for squat weight. It just so happens that these two are the same height and have the exact same weight, but the center of their weight is not in the same part of their body. This is because the girl has much more weight in the top half of her body and the boy has more weight in the bottom half. This difference in center of mass is a direct result of the different distributions of mass throughout both of their bodies. Moments and Mass There are two main components to finding the center of mass of an object. The first, unsurprisingly, is the mass of the whole object. In this case of the boat example, the mass will be uniform throughout the entire object. This is ideal a majority of the time as it drastically reduces the difficulty...
Post a Comment
Read more