The overall difference in typical image quality between a current smartphone like the Huawei P30 Pro and an older DSLR like the Canon EOS Rebel T1i/500D isn't the difference between the sensors. The difference is about who makes the decisions when shooting and, more importantly, in post processing about how both are done.

Recent smartphones have gotten very good at computational photography. Sometimes they even take multiple exposures and use AI to either select the best of several frames or computationally combine the multiple frames into a single image. To use an older, or even more current, DSLR or mirrorless interchangeable lens camera to do the same type of computational photography requires the user to do things and make decisions that the smartphones are automatically doing under the surface.

One must decide, for example, how many frames to take in quick succession. Should they all be at the same exposure level or bracketed? Should they all be at the same focus distance or bracketed? Should exposure be based upon the brightest highlights in the scene? Or the darkest shadows? The options are multiplied even more when processing the image data from the sensor to produce a viewable image. A highly skilled post processor can get a lot out of even fairly poor, by 2019 standards, sensors if the image was exposed properly. Learning how to expose properly in challenging light is an art that takes a lot of time and effort to learn. It also takes a lot of knowledge, skill, and experience to be able to post process better than modern AI routines included in smartphones and, to a lesser extent, digital cameras.

For most folks who only do casual photography, it's not worth the time and effort to learn how to do it themselves with a dedicated camera instead of letting the programmed routines of a smartphone do it for them.