Error in using R-CNN Network to detect an object.

I am working with R-CNN to identify an arrow. It works with one R-CNN network and when i try to run the same code with another R-CNN trained network it shows the following error.

In the following picture shows the i code i am using.

In the code, in the detect function if i use "rcnnArrow" , a R-CNN trained Network, there is no error. But if i use "rcnndifferentarrows1" another trained R-CNN network, it shows the error in the first picture. In the figure below shows the two variables in the workspace.

How can i rectify this problem?

[Ajay Pattassery]

uint8 is a valid image type for an R-CNN object detector.

1. Can you attach the rcnndifferentarrows1 detector?
2. Also, paste the complete error output as it is. (I guess the above is a part of error)

[Philip Kanjirathingal Kurian]

1. No, i cant attach the rcnndifferentarrows1 detector, because its a huge file to attach here.
2. The erroe which appears when i run the program has been pasted in the question. Thats the whole part of the error. There are no other errors displayed.

[Ajay Pattassery]

can you put a break point on the following line

 [Bbox, Score, Labels]= detect(rcnndifferentarrows1, image)

and paste the output of

which detect
whos image
disp(rcnndifferentarrows1)