The CoCr2O4 is multiferroic spinel and crystallizes in the normal cubic F d¯3m structure with Co2+ (electronic configuration 3d7, spin S = 3/2) and Cr3+ions (3d3, S = 3/2) occupying tetrahedral and octahedral sites, respectively. Multiferroicity in CoCr2O4 is related to collinear ferrimagnetic spin order sets in below TC ≈ 95 K, followed by a transition into an incommensurate long-range conical-spiral state at Ts ≈ 26 K. In my opinion the increase of Ms in abovementioned temperature range is associated with presence of magnetodielectric effect and changes of magnetostriction vs temperature wich affect to lattice parameters of the crystal as well as to spin configurations.

@Ghazanfar Mehboob

Interesting question!!! Could you please mention the magnetic field you used for Ms measurement?? Magnetic field is very important in Ms vs T measurment.

what  is the the temperature range you used?? and the if possible you can provide the graph for better answer.

I found M-H loop at different temperature 5 K, 25 K 50 k and 75 K. And field range for all loop 0 to 5000 Oe

For any given particle size, increasing the temperature reduces magnetization. This happens because, as temperature increases, more thermal energy is supplied and individual electron spins become more likely to be in higher energy states, pointing opposite to their neighbors, and less lined up, leading to a reduction in the total magnetism. Because of this a higher magnetic field is required to saturate the sample.

Additionally, for any particle size, increasing the temperature reduces the coercivity. This occurs because as temperature increases, more thermal energy is supplied and individual electron spins become more likely to be in higher energy states, pointing randomly, opposite to their neighbors and less aligned, so a smaller field is required to reduce remnant magnetization to zero, leading to a reduction in coercivity.

Please read the following paper for additional insight: 

http://aip.scitation.org/doi/full/10.1063/1.4975695

In a magnetostrictive material, the saturation magnetization can change with changes in the lattice parameters (size) of unit cell that can cause  changes in the orientation of spins and hence their mutual interaction.  

In ferrimagnetic systems, there is a possibility that spontaneous magnetization increases with temperature up to Tc , in particularly Neel P, M, L and N ferrimagnetic types (see L. Néel, Ann. Géophys.  5 (1949) 99-136)

@Ghazanfar Mehboob

It seems that my doubt is correct!!

You filed scan range is small for this type of system. 

Actually if you are in the situation where you have not make you system ACTUALLY saturated, you will definitely get the Ms value(not at all). Please do not confuse your somewhat looking like a linear type region in your MH is not linear actually(see by zooming the saturated region).

I find one reference for e.g. Journal of Applied Physics 115(12):123908-123908-10

see Figure 6 and specially FC regime. If you are working in non saturated regime(seems to be your case) then definitely your Ms will be looking like increasing with Temp.

Be careful during Ms calculation. By increasing the Temp the Ms will always decreases as explained by  Mr Amir Aslani above.

In any system (to the best of my knowledge) the Ms will never increase with T. If increase first we should doubt our interpretation and then the measurement. It all things are sorted out then we can reach to the goal.

So I can tell you step to resolve your problem if you (really) want ;

1. Do your MH in at least +-5T (If you have). If not?

2. Set highest field in your system and cool down the system to lowest T and then measure the Ms in the heating mode. You Ms should decrease with increasing T if there is not any AFM transition in your system on the way.

So moral is you need high magnetic field to resolve this problem.

I hope you can resolve it now!!

Best Wishes.