Update image_cluster_extractor.py

**Identifying multiple objects ** calculates the areas of each contour in a list of contours, filters out contours with areas smaller than a specified threshold, sorts the remaining contours by area in descending order, selects the largest num_of_objects contours, and stores them in a new list. Verify that the selected contours are actually objects checks if the number of points in each contour's associated point cloud cloud.points is greater than or equal to a specified threshold self.descriptor.parameters.min_points_threshold.By verifying that the selected contours have a sufficient number of points, we can make sure that we are identifying true objects in the point cloud data.