Proposed interval type II intuitionistic fuzzy c means with spatial triangular fuzzy number

This algorithm decreases various uncertainties in histopathology images. It comprises the following steps:

• Step 1: Set the initial values for the centroids γ_i wherever i =1 to c.
• Step 2: The primary memberships of higher μ¯ik and lesser membership of μ_ik are set according to two fuzzifiers m₁ and m₂. The constant m satisfies the condition m₁, m₂ ≥ 1. The two fuzzifiers help to construct footprint of uncertainty{Jm1(U,v)=∑j=1N∑i=1C(μjk)m1dik2Jm2(U,V)=∑j=1N∑i=1C(μjk)m2dik2(3.3)
• Step 3: The k represents the Euclidian distance dik2 between pixel samples; the rate of spatial order is computed for every SP_ij pixel as follows:SPA¯ik=∑j=1Nμij¯(dkj)−1∑j=1N(dkj)−1(3.4)SPAik_=∑j=1Nμij_(dkj)−1∑j=1N(dkj)−1(3.5)The value of the spatial information is defuzzified asSPAik=(SPAik¯+SPAik)/2_(3.6)
• Step 4: Compute the matrix of membership grades U_ik by means of the distance calculation asNewDistjk=||xk−γi||2.(1−αe−SPik)(3.7)
• Step 5: Update the centroid of clusters asDDj=[γ1j,γ2j,…,γcj](3.8)
• Step 6: Lastly, verify the stop condition, if max (|J^(j+1) − J^(j)|), go to subsequent process otherwise go to step 2.

This proposed algorithm detects the weed from various crop images by selecting GLCM features from segmented image. Segmentation of the weed from crops and soil in the input image properly. These results are shown in Figure 3.3. The weed and crops in same color and size, so machine learning techniques are applied to detect and separate weeds and plant areas. The results guide the automatic sprayer and weed removal robotics. This proposed fuzzy c means algorithm improves the results and accurately detects weed areas. This process helps to reduce the manual work and improve farming process.