How to analyze the zip file you just downloaded?
Within the zip file, you will find 7 types of files.
1. classes.txt : The classes of objects that can be detected by this object detection model
2. count_results.csv : The number of snails detected by the model, as well as the average bounding box pixel size and blurriness
3. detection_results.csv : The detailed information of the detected bounding box(es)
4. results.txt : The raw results with every detailed information written in .json format
5. image.jpg/.png/.jpeg/.gif : The original image that you uploaded
6. image.txt : The detection result of the detected image
7. image_detection.jpg/.png/.jpeg/.gif : The image with detection drawn on top of your uploaded image(s)
While most of the documents in the zip file are self-explanatory, the classes.txt and image.txt are in a specific format for the use of object detection.
In the classes.txt, each row represents one class of objects. In our current model, we can identify 7 classes of objects, namely, Biomphalaria, Radix, Pool, Pool_triangle, Pool_spiral, Pool_oval, Gyraulus. The order of classes is important to allow correct identification of the detection.
The image.txt is formatted in YOLO format with each row representing one detection (bounding box) of a snail.
The first column is ranged from 0 to 6 in which each number refers to the number of rows in the classes.txt. For example, Biomphalaria is represented by "0" as it is the first row of classes in the classes.txt file.
The second and third columns are the center coordinates of the detection, normalized to the size of the image.
The fourth and fifth columns are the height and width of the detection, normalized to the size of the image.
This API is financed by the ATRAP project of the Development Cooperation program of the Royal Museum for Central Africa with support of the Directorate-General Development Cooperation and Humanitarian Aid. The ATRAP project is coordinated by the Royal Museum of Central Africa (Belgium) with the collaboration with the University of Kinshasa (Democratic Republic of Congo), Mabarara University of Science and Technology (Uganda) and KU Leuven (Belgium).
