2. Consider the image in canvas \'combined_lanes jpg\'. This image was obtained

Question

2. Consider the image in canvas 'combined_lanes jpg'. This image was obtained from the TurtleBota ra (focal length 3.6mm, and pixel size 1.12 m x 1.12 m). Burger robot with a Raspberry Pi came The ground plane/floor ends at row 1150 in the image - meaning all pixels above that row are guaranteed to be above the image plane. Use Canny or Sobel or Prewitt edge detector in Matlab to detect the white lines on the ground plane/floor How far away in distance is the bottom of the V corresponding to the intersection of the two white lines -camera's true origin is center row (1232) and center column (1640) of the image. You can assume that the rows and columns of the image are inverted with respect to the pin-hole (perspective) projection - both with respect to up/down and right/left - a feature of many imaging systems. a. 20 Pts b. 20 Pts]

Explanation / Answer

BW = edge(I)

BW = edge(I,'Sobel')

BW = edge(I,'Sobel',threshold)

BW = edge(I,'Sobel',threshold,direction)

BW = edge(I,'Sobel',threshold,direction,'nothinning')

[BW,threshOut] = edge(I,'Sobel',___)

BW = edge(I,'Prewitt')

BW = edge(I,'Prewitt',threshold)

BW = edge(I,'Prewitt',threshold,direction)

BW = edge(I,'Prewitt',threshold,direction,'nothinning')

[BW,threshOut] = edge(I,'Prewitt',___)

BW = edge(I,'Roberts')

BW = edge(I,'Roberts',threshold)

BW = edge(I,'Roberts',threshold,'nothinning')

[BW,threshOut] = edge(I,'Roberts',threshold,'nothinning')

BW = edge(I,'log')

BW = edge(I,'log',threshold)

BW = edge(I,'log',threshold,sigma)

[BW,threshOut] = edge(I,'log',___)

BW = edge(I,'zerocross',threshold,h)

[BW,threshOut] = edge(I,'zerocross',___)

BW = edge(I,'Canny')

BW = edge(I,'Canny',threshold)

BW = edge(I,'Canny',threshold,sigma)

[BW,threshOut] = edge(I,'Canny',___)

BW = edge(I,'approxcanny')

BW = edge(I,'approxcanny',threshold)

[gpuarrayBW,threshOut] = edge(gpuarrayI,___)

Depiction

BW = edge(I) restores a paired picture BW containing 1s where the capacity discovers edges in the information picture I and 0s somewhere else. The info picture I is a force or a double picture. BW is an indistinguishable size from I.

Of course, edge utilizes the Sobel edge recognition technique, however you can indicate any of these different strategies: Canny (or a Canny guess), Laplacian of Gaussian (log), Prewitt, Roberts, or Zero-intersections. The parameters you determine fluctuate contingent upon the technique you pick. The accompanying area detail the parameters upheld by every technique.

BW = edge(I,'Sobel') distinguish edges utilizing the Sobel technique. This technique discovers edges utilizing the Sobel estimate to the subordinate. It returns edges at those focuses where the slope of I is greatest.

BW = edge(I,'Sobel',threshold) restore all edges that are more grounded than edge. On the off chance that you don't determine edge, or is you indicate purge sections ([]), edge picks the esteem consequently.

BW = edge(I,'Sobel',threshold,direction) indicate the course in which the capacity searches for edges in the picture: 'flat', 'vertical', or 'both'.

BW = edge(I,'Sobel',threshold,direction,'nothinning') indicate whether to skirt the extra edge-diminishing stage, 'nothinning'. Avoiding this stage can enhance execution. The default esteem is 'diminishing'.

[BW,threshOut] = edge(I,'Sobel',___) restores the limit esteem.

BW = edge(I,'Prewitt') recognize edges utilizing the Prewitt strategy. This technique discovers edges utilizing the Prewitt estimation to the subsidiary. It returns edges at those focuses where the slope of I is greatest.

BW = edge(I,'Prewitt',threshold) restore all edges that are more grounded than edge. In the event that you don't indicate edge, or is you determine purge sections ([]), edge picks the esteem consequently.

BW = edge(I,'Prewitt',threshold,direction) determine the bearing in which the capacity searches for edges in the picture: 'even', 'vertical', or 'both'.

BW = edge(I,'Prewitt',threshold,direction,'nothinning') determine whether to avoid the extra edge-diminishing stage, 'nothinning'. Skirting this stage can enhance execution. The default esteem is 'diminishing'.

[BW,threshOut] = edge(I,'Prewitt',___) restores the edge esteem.

BW = edge(I,'Roberts') identify edges utilizing the Roberts technique. This strategy discovers edges utilizing the Roberts guess to the subordinate. It returns edges at those focuses where the inclination of I is greatest.

BW = edge(I,'Roberts',threshold) restore all edges that are more grounded than edge. On the off chance that you don't determine limit, or is you indicate exhaust sections ([]), edge picks the esteem consequently.

BW = edge(I,'Roberts',threshold,'nothinning') indicate whether to skirt the extra edge-diminishing stage, 'nothinning'. Avoiding this stage can enhance execution. The default esteem is 'diminishing'.

[BW,threshOut] = edge(I,'Roberts',threshold,'nothinning') restores the edge esteem.

BW = edge(I,'log') identify edges utilizing the Laplacian of Gaussian ('log') technique. This strategy discovers edges by searching for zero-intersections in the wake of sifting I with a Laplacian of Gaussian channel.

BW = edge(I,'log',threshold) restore all edges that are more grounded than limit. In the event that you don't determine limit, or is you indicate discharge sections ([]), edge picks the esteem consequently.

BW = edge(I,'log',threshold,sigma) determine sigma, the standard deviation of the 'log' channel. The default sigma is 2; the measure of the channel is n-by-n, where n=ceil(sigma*3)*2+1.

[BW,threshOut] = edge(I,'log',___) restores the limit esteem.

BW = edge(I,'zerocross',threshold,h) distinguish edges utilizing the 'zerocross' strategy. This technique discovers edges by searching for zero-intersections in the wake of sifting I with a channel that you determine, h. The edge work returns edges that are more grounded than edge. In the event that you don't indicate edge, or is you determine exhaust sections ([]), edge picks the limit esteem naturally.

[BW,threshOut] = edge(I,'zerocross',___) restores the limit esteem.

BW = edge(I,'Canny') distinguish edges utilizing the Canny strategy. The Canny technique discovers edges by searching for neighborhood maxima of the angle of I. The edge work computes the slope utilizing the subsidiary of a Gaussian channel. This strategy utilizes two edges to identify solid and feeble edges, incorporating frail edges in the yield on the off chance that they are associated with solid edges. By utilizing two limits, the Canny technique is more outlandish than alternate strategies to be tricked by commotion, and more inclined to recognize genuine powerless edges.

The Canny technique isn't upheld on a GPU.

BW = edge(I,'Canny',threshold) restore all edges that are more grounded than edge. In the event that you don't indicate edge, or in the event that you determine exhaust sections ([]), edge picks the esteem naturally. limit is a two-component vector in which the primary component is the low edge, and the second component is the high edge. On the off chance that you determine a scalar, edge utilizes this incentive for the high esteem and uses threshold*0.4 for the low limit.

BW = edge(I,'Canny',threshold,sigma) indicate sigma, the standard deviation of the Gaussian channel. The default sigma is sqrt(2). edge picks the measure of the channel consequently, in light of sigma.

[BW,threshOut] = edge(I,'Canny',___) restores the limit esteems as a two-component vector.

BW = edge(I,'approxcanny') recognize edges utilizing the estimated Canny strategy. The 'approxcanny' technique is an inexact variant of the Canny edge discovery calculation that gives speedier execution time to the detriment of less exact recognition. For the approxcanny technique, skimming point pictures are relied upon to be standardized in the range [0 1].

The surmised Canny technique isn't upheld on a GPU.

BW = edge(I,'approxcanny',threshold) indicates affectability edges for the 'approxcanny' strategy. limit is a two component vector, the primary component of which determines the lower edge for edge quality, beneath which all edges are dismissed. The second component indicates the higher edge, above which all edge pixels are protected. The scope of qualities permitted is between [0 1]. In the event that you indicate a scalar, edge utilizes this incentive for the high esteem and uses threshold*0.4 for the low limit. On the off chance that you don't determine limit, or if edge is vacant ([]), edge picks low and high esteems naturally.

Related Questions

Navigate

• Browse (All)
• Browse 2
• Subjects

---

---

Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.