Embedded Files
Research
Research
This automated waste sorter machine has a tube like top where waste is inserted then there is some sort of sensor in there that detects what type of waste it is then in the bottom half the re are rotating bins that move based on the output of the sensor.
This waste sorting machine has a entrance hole at the top for waste, there a sensor detects what kind of waste it is then the bottom rotates to a specific side and drops the waste.
This sorting machine used a conveyer belt and different dividers that pushed the waste in the correct bin. the sensors seem to be on the side of the conveyer
stepper motor moving
stepper motor moving
today in class we where introduced to a new motor it turns similar to a servo but its more exact and has a wider range to spin.
recycling machine learning model (final project)
recycling machine learning model (final project)
What is machine learning?
What is machine learning?
Machine learning is a type of artificial intelligence were machines should be able to learn and adapt through experience like training.
The goal of my model
The goal of my model
The ultimate goal of my machine learning model is to distinctly identify four different recyclable items with 90% accuracy or more .
Bottle
fork
cereal box
juice cup
Data collecion process
Data collecion process
I used 250 pictures of each item for my data the pictures where 48x48 pixels.
bottle
fork
cereal box
juice cup
My model
My model
Here is the link to my final model.
This model first connects to my google drive from there it has access to my data which is stored in their own individual folders relating to the item. From there I wrote the code for my data to be reshaped into a pixel format that computer can understand. After that I had to test and train my code to understand the data input. Then I evaluated the code to see its accuracy. And wrote a code to see any errors my model made.
My data
My data
Here is the link to the data I collected for my model. To collect this I had to run a code on Mrs.Whites computer and constantly rotate the item while the computer took pictures repeatedly.
Building the model
Building the model
Accurately labled images
Accurately labled images
The 'prediction:' label is what my model came up with and the 'true recycle:' is where I have the picture in my google drive which I organized accurately.
bottle
bottle
cereal box
cereal box
fork
fork
Mislabled images
Mislabled images
Since my model had 100% accuracy there where no mislabeled images.
Page updated
Report abuse