Picture categorization is a fundamental task in vision recognition that aims to understand
and label an image in its entirety. While object detection works with the categorization and
placement of many elements inside an image, image classification often pertains to photographs
containing a single object. The development of sophisticated parallel computers in tandem with
the introduction of contemporary remote sensors has fundamentally changed the picture categorization
theory. Various algorithms have been created to recognise objects of interest in pictures and
then categorise them and practise. In recent years, a number of authors have offered a range of
classification strategies. However, there are not many studies or comparisons of classification
techniques in soft computing settings. These days, the use of soft computing techniques has improved
the performance of classification methods. This work explores the use of soft computing
for image classification for various applications. The study explores further details regarding new
applications and various classification technique types. To promote greater study in this field, important
problems and viable fixes for applications based on soft computing are also covered. As a
result, researchers will find this survey study useful in implementing an optimal categorization
method for multiple applications.