Biospeckle activity measurement of Indian fruits using the methods of cross-correlation and inertia moments

Biospeckle activity measurement of Indian fruits using the methods of cross-correlation and inertia moments

Optik 124 (2013) 2180–2186 Contents lists available at SciVerse ScienceDirect Optik journal homepage: www.elsevier.de/ijleo Biospeckle activity mea...

1MB Sizes 9 Downloads 34 Views

Optik 124 (2013) 2180–2186

Contents lists available at SciVerse ScienceDirect

Optik journal homepage: www.elsevier.de/ijleo

Biospeckle activity measurement of Indian fruits using the methods of cross-correlation and inertia moments M.D. Zaheer Ansari, Anil Kumar Nirala ∗ Biomedical Physics Lab, Department of Applied Physics, Indian School of Mines, Dhanbad, India

a r t i c l e

i n f o

Article history: Received 8 February 2012 Accepted 22 June 2012

Keywords: Biospeckle Shelf-life Cross-correlation coefficient Respiration Co-occurrence matrix

a b s t r a c t This paper presents biospeckle activity evaluation using two methods namely spatial–temporal speckle correlation and inertia moment applied for three different Indian fruits namely apple, pear and tomato for the first time. The bioactivity was determined by means of the cross-correlation functions of the intensity fluctuations and using inertia moment of the THSP image of biospeckles. Significant changes in bioactivity were observed during their shelf lives. From the study, it is found that the activity is higher for pear in comparison to the apple and tomato as predicted by IM method. Biospeckle activity decreases with aging of the fruits but the decrease is more in pear & relatively less in the case of apple and tomato as predicted by cross correlation technique. Further it is also concluded that the activity changes according to their respiration rates. By the comparative study between the two methods it is found that IM is more reliable to predict the bioactivity levels in fruits in comparison to the cross-correlation technique as IM measures the bioactivity directly. © 2012 Elsevier GmbH. All rights reserved.

1. Introduction Due to the socioeconomic importance of guaranteeing quality food, there is a persistent search for the improvement of food products in terms of safety, health, appearance, and many other market attributes. So there is need to evaluate fruits quality at different stages of pre- and post-harvest technology in order to provide product of the best quality to consumers [1]. Recently, a few interesting optical techniques and devices have been developed and successfully used for nondestructive evaluation of fruit and vegetables: vis/NIR spectrophotometry [2], time-resolved reflectance spectroscopy [3], hyperspectral backscattering imaging [4,5], laserinduced light backscattering [6,7] or chlorophyll fluorescence [8–10]. Nikolai et al. [11] have reviewed most of the above techniques, collectively naming them NIR spectroscopy. Biospeckle is a phenomenon that occurs when laser light reaches an object that exhibits some kind of activity or dynamic process that can be biological or non biological. It is an optical technique for nondestructive evaluation of biological materials. In the method, coherent laser light illuminates an object to be investigated. The backscattered light interferes and a speckle pattern is created in an observation plane. If the sample does not show activity, the speckle

∗ Corresponding author. Tel.: +91 3262235483; fax: +91 3262296563. E-mail addresses: [email protected] (M.D.Z. Ansari), [email protected], [email protected] (A.K. Nirala). 0030-4026/$ – see front matter © 2012 Elsevier GmbH. All rights reserved. http://dx.doi.org/10.1016/j.ijleo.2012.06.081

pattern is stable in time. However in the case of biological samples, the speckle pattern consists of two components: the static one from stationary elements of the tissue and the variable one from moving particles of the tissue. The variable in time speckle pattern is characteristic for biological tissue and has been called as the biospeckle. When the speckle is formed by the portion of light scattered by movable elements, it is modulated by their movement. Then, it is difficult to identify precisely which element is responsible for the scattering, because a laser photon can penetrate the vegetal material and suffer multiple deviations in its path before it eventually returns to the surface and reaches the light detector. Cellular structure may vary from one specimen to another, and the movement of cell components can also vary. This movement is also modified by the age of the cell. Therefore, it can be expected that the speckle formed by different cells is different; in addition, the speckle will change as the cell ages. In other words, a speckle can be used to distinguish specimens and also the age of biological material. We consider the age of a fruit as the number of days passed since the sample was brought from the mother plant. Temporal analysis of speckle can be used to estimate the age of fruits after harvest, assuming the correlation function as a sort of parameter to classify differences in speckle. Bragga et al. [12] have shown that processes related with movement of the scattering centers in the tissue, such as cytoplasmic streaming, organelle movement, cell growth and division during fruits maturation and biochemical reactions are responsible for a

M.D.Z. Ansari, A.K. Nirala / Optik 124 (2013) 2180–2186

2181

certain biospeckle activity. Brownian motions should be considered as a source of biospeckle activity too. It has also been shown that biospeckle activity changes with an age or with some surface properties, for example an infection of a biological object. It is also found that biospeckle activity changes with water [13], chlorophyll and starch contents [14,15]. Less chlorophyll content causes higher apparent biospeckle activity [15] due to light absorption by this pigment and in consequence shallower light penetration through a tissue. So far, attempts to apply biospeckle methods in biological studies include measurements of blood flow in blood vessels [16], viability of seeds [17,18], activity of parasites in living tissues [19,20], analysis of maturation and bruising of fruits and vegetables [21,22]. These studies showed that decaying of a tissue conditions caused by age, illness/infection or damage, relates with lower biospeckle activity. In this paper two different methods known as spatial–temporal speckle correlation technique and Inertia Moment have been utilized to interpret biospeckle data of three different Indian fruits namely apple, pear and tomato. The methods are noncontact and non-destructive and have been used for the bioactivity evaluation of fruits during their shelf-lives.

Table 1 Bioactivity of different Indian fruits during their shelf lives.

2. Biospeckle cross-correlation method

C k

Spatial–temporal speckle correlation (DSC) technique is a measuring process based on the correlation analysis of a reference speckle pattern of the specimen in its initial state with sequential speckle patterns while changing the surface or subsurface of the specimen [23]. In order to obtain the temporal dependencies of biospeckle pattern movement speed, each pattern is separated on M by N sub images and each (m, n)th subimage is correlated with the respective subimage belonging to any other pattern of the same studied area. The cross-correlation function of respective fragment pairs with identical indexing can be expressed as [24]: 1  Cm,n (k, l) = rm,n (i, j)sm,n (i + k, j + l) IJ I

J

i

j

(1)

where rm,n is the (m, n)th fragment of the pattern 1, sm,n is the (m, n)th fragment of the pattern 2, m = 1,. . .,M and n = 1,. . .,N are the numbers of fragments, i = 1,. . ., I and j = 1,. . .,J are the numbers of fragment pixels, k = 1,. . .,K and l = 1,. . .,L are the discrete samples of the cross-correlation function. To obtain the temporal dependencies of biospeckle pattern movement speed, each pattern was separated on M by N sub images and each (m, n)th sub image was correlated with respective sub image belonging to any other pattern of the same studied area. As a result, cross-correlation coefficients were obtained using the equation:

        t +k t +k  St0 − St0  Si,j0 − Si,j0   i,j i,j k   Cm,n =   t0 t0 +k i,j i,j  

(2)

where i, j is the pixel number in the (m,n)th sub image of the digital biospeckle pattern, i = 1,. . .,I; j = 1,. . .,J; m = 1,. . .,M; n = 1,. . .,N; Si,j is the i, jth pixel intensity, k is the number of biospeckle patterns,  is the interval between two  adjacent frames containing recorded biospeckle patterns, i,j =





Si,j − Si,j

2 

is the vari-

ance. Calculation of the cross-correlation coefficients for series of speckle pattern’s sub images recorded in the given temporal order allows receiving the temporal dependencies of these coefficients as

Fruit commodity

Shelf life (days)

BA (average)

Apple

1 2 4 6

0.523 0.476 0.454 0.426

Pear

1 2 4 6

0.617 0.591 0.521 0.478

Tomato

1 2 4 6

0.081 0.076 0.068 0.061

BA, bioactivity.

functions of the biospeckle pattern movement speed. Due to homogeneity of biospeckle properties of each surface fragment, intensity of which is calculated as a mean value of intensities of all correlation peaks and the correlation coefficient can be expressed as:

        t0 +k t0 +k  St0 − St0  Sim,jn − Sim,jn   im,jn im,jn  =   t0 t0 +k im,jn im,jn  

(3)

where im = 1,. . .I,. . .,2I,. . .,MI and jn = 1,. . .,J,. . .,2J,. . .,NJ. 2.1. Experimental In order to study the biospeckle temporal properties of the fruits experimental setup was mounted. An expended He–Ne laser (2 mW) with  = 632.8 nm, beam was used for recording the biospeckle patterns of the fruits with a CCD camera connected to PC. The optical part of the setup was kept on the vibration free table for decreasing the influence of external perturbations. For the experiment, the recording time was equal to 15 s with the frame rate of 20 fps. The observation area was marked on each of them. Measurements were performed every day on the same places of the fruits. The images of speckles were obtained as a movie. The movies were then recorded as speckle pattern series (series of images) and they were processed using PC by special code developed for this purpose. The code calculated the correlation coefficients Ck according to Eq. (3). These data were used to plot the temporal dependencies of these coefficients. We planned our study as follows-we selected three types of different climacteric (with starch reserves) Indian fruits namely apple, pear and tomato. The fruits were fresh and conditioned at room temperature for one day before a shelf life program consisting of 1, 2, 3, 4, 5, 6 and 7 days of storage. We have plotted cross correlation coefficients for all the three fruits on inter-day basis. We have taken a series of observations for the fruits and repeated these observations three times almost in identical situation with temperature in the range of 24–27 ◦ C and humidity in the range of 62–65%. 2.2. Results Biospeckle activity (BA) was calculated using the correlation coefficient Ck , where k = 0, 1, 2, 3,. . .and  = 1/15 s. Ck was calculated as the correlation coefficient of data matrix of the first frame (k = 0) with the data matrixes of the following frames (at k) from the bitmaps of the biospeckle. In this study, 14 C was analyzed only as the correlation coefficient between the first frame k = 0 and the frame at k = 14 s. Then, biospeckle activity BA = 1 − 14 C value

2182

M.D.Z. Ansari, A.K. Nirala / Optik 124 (2013) 2180–2186

Fig. 4. Ck as a function of shelflife storage day for pear. Fig. 1. Temporal changes of crosscorrelation coefficients during shelf-life storage for apple.

Fig. 5. Temporal changes of cross correlation coefficient during shelf-life storage for tomato. Fig. 2. Ck as a function of shelflife storage day for apple.

Fig. 6. Ck as a function of shelflife storage day for tomato. Table 2 Mean rate of decrease of BA with commodities. Fig. 3. Temporal changes of crosscorrelation coefficient during shelf-life storage for pear.

was determined. Here we are presenting the representative typical graphs for the fruits (Figs. 1–6). Table 1 shows that BA (bioactivity) for different fruits decrease during their storage period. The change is gradual and significant. The decrease is fast in pear and is comparatively lesser in the case of apple and tomato (Table 2).

Commodity

Mean rate of decrease of BA

Pear Apple Tomato

0.044 0.031 0.004

3. Methods of inertia moment [25] The temporal variations of biospeckle are studied also using the technique known as inertia moment. The method has input data

M.D.Z. Ansari, A.K. Nirala / Optik 124 (2013) 2180–2186

as time history of the speckle pattern (THSP), which is an image that represents the variation, along the time, of a region from the biospeckle of the material illuminated with laser beam. It has been used as a reliable technique for quantifying the THSP, returning a number that indicates low or high activity from the materials monitored. The analysis of the activity in a speckle pattern produced by the laser illumination of the fruit is conveniently evaluated by using the display known as THSP [26]. For every state of the phenomenon being assessed; successive images of the dynamic speckle pattern are registered. A certain column (for example, the middle one) is selected in each of them. Then, a new image is constructed by setting, side by side, the chosen column extracted from the successive patterns. The resulting image is named ‘time history of the speckle pattern (THSP)’. Its rows represent different points on the speckle pattern and the columns their intensity in a sequence of regularly spaced time steps. The activity of the sample appears as intensity changes in the horizontal direction, that is, along the rows. 3.1. Formation of co-occurrence matrix (COM) from THSP A co-occurrence matrix (COM) expresses the number of transitions of each THSP pixel with respect to its immediate neighbor. In biospeckle analysis, the COM represents an intermediary step to obtain the inertia moment (IM) of a sample. The co-occurrence matrix is defined as: MCO = [Nij ]

(4)

where the entries are the number (N) of occurrences of a certain intensity value i, that is immediately followed by an intensity value j. In the present work, the variable of interest is time. Then the involved N values are the occurrences of a certain grey value i followed in the next time step by a value j in the THSP as described above. When the intensity does not change, the only non-zero values of this matrix belong to the principal diagonal. As the sample shows activity, intensity values change in time, the number (N) outside the diagonal increases and the matrix resembles a cloud. Nevertheless, this matrix is sparse; it is mostly composed by zero values. 3.2. Generation of inertia moment (IM) of the COM One of method for analyzing the THSP is the use of the inertia moment (IM) [27] of the COM. Inertia moment (IM) is based on occurrence of successive pixels intensity values from a THSP image for generating the number that quantify activity. From THSP, inertia moment uses a co-occurrence matrix (COM). This co-occurrence matrix is represented with 8 bits in grey scale, where pixels intensity varies from 0 (black) until 255 (white). Because of this, the co-occurrence matrix is 256 × 256 positions. Each position of this co-occurrence matrix stores the number of occurrences of a grey scale intensity i that is followed in the next time by a grey scale intensity j, where i and j vary from 0 to 255. Low activity concentrates values in main diagonal (lighter spots) and high activity has values distributed at all positions of the matrix, not only in the center. Since values not null out of the main diagonal of the co-occurrence matrix are an evidence of higher activities, inertia moment will implement a second order moment to quantify the biospeckle, as follows [27]: MI =



Mij (i − j)

2

2183

where Mij =

N

ij j

Nij

In this way, values out of the main diagonal contribute more for IM than the values nearest the main diagonal, being possible to obtain a number that represents high or low activity. 3.3. Experimental For the experiment apple, pear and tomato were taken as fresh as possible and stored in a cool place. The observation area was marked on each of them. 225 images were taken at an interval of 1 s to find the IM against its activity values. Also 200 images were taken at an interval of 1 min to find the variation of IM value against time. Successive images were registered, digitized to 8 bits (256 grey levels) and stored in image processing system. Measurements were performed on the same place of the samples. For the co-occurrence matrix analysis, a column of the free propagation speckle pattern was read every 1 s for the first result and every 1 min for the second result and then, a composite image of 225 by 225 pixels was formed by stacking consecutive columns. Finally, this image was retrieved and the second-order moments of its co-occurrence matrix were calculated. The speckle images were then registered, the THSP was constructed and the IM was calculated. The whole set of reading was repeated three times and for each of them IM was calculated. We selected three types of Indian fruits namely apple, pear and tomato. We have taken a series of observations for the fruits and repeated these observations three times almost in identical situation with temperature in the range of 20–25 ◦ C and humidity in the range of 55–60%. We are presenting the representative typical graphs for the three fruits. 3.4. Results Figs. 7–9 show the co-occurrence matrices of apple, pear and tomato for different days respectively for which we can consider it to have high, intermediate and low activities respectively. It can be seen that the points in the main diagonal represent no change of intensity while the spread of points out of the diagonal represents time intensity changes. So, if the activity of the biological tissue is low, intensity changes are slow and the only appreciable values of the matrix appear near to the diagonal. Conversely, if activity is high, the fast intensity changes produce high values far away

Fig. 7. Co-occurrence matrix MCO for 1st, 2nd, 4th and 6th day of pear.

(5)

ij

Fig. 8. Co-occurrence matrix MCO for 1st, 2nd, 4th and 6th day of apple.

2184

M.D.Z. Ansari, A.K. Nirala / Optik 124 (2013) 2180–2186

Fig. 9. Co-occurrence matrix MCO for 1st, 2nd, 4th and 6th day of tomato.

Fig. 12. Variation of IM with frequency for tomato.

Fig. 10. Variation of IM with frequency for pear.

Fig. 13. Variation of IM with time (s) for different fruits. Table 3 Variation of IM values with frequency.

Fig. 11. Variation of IM with frequency for apple.

the principal diagonal of the matrix. Hence the activity goes on decreasing with the aging of the fruits. 3.5. Variation of IM with activity On the first day when the fruits were fresh IM value increases rapidly which gives the variation of pixel or intensity value but when the activity of biological sample goes on decreasing with time the IM value decreases (Figs. 10–13). Table 3 shows the variation of IM (max.) with the frequency. The variation of frequency implies the variation of pixel values in the THSP. Low frequency means low variation in the pixels of particular THSP image while there is a large pixel variation for higher frequencies. As can be seen from Table 3 with the changing frequency there is a comparative change in the IM values of the fruits. Thus there are higher variations in the pixels of THSP image for pear in comparison to the apple and tomato and simultaneously the IM values.

Shelf-life days

Max. IM values

frequency levels

Pear 1 2 4 6

3242 3522 3758 4834

3965 4223 4761 6084

Apple 1 2 4 6

3155 3289 3519 4028

2945 3721 4223 5903

Tomato 1 2 4 6

2833 3144 3393 3880

3079 3665 3931 4135

So the bioactivity as represented by non dimensional IM values is higher for pear in comparison to apple and tomato. Table 4 shows the variation of IM values with time. The biospeckle activity for different fruits decreases with the aging of the sample. 4. Discussion Laser light falling on the fruit surface is elastically scattered on each boundary, like intracellular membranes, organelles and other particles. If any object due to cyclosis is moving, the scattering

M.D.Z. Ansari, A.K. Nirala / Optik 124 (2013) 2180–2186 Table 4 Variation of IM values with Time. Pear

Apple

Tomato

IM value (×106 )

Time (s)

IM value (×106 )

Time (s)

IM value (×106 )

Time (s)

9.06 7.42 5.76 5.17 4.82 4.63

20 40 80 100 120 140

8.02 6.31 6.28 5.93 5.48 5.67

20 80 100 120 140 160

5.87 5.46 5.29 4.85 4.21 4.16

20 40 60 80 100 120

causes an unstable biospeckle pattern. So the biospeckle activity is a function of particles activity (mobility) and vitality of a tissue. Climacteric fruits like apple, pear and tomato accumulate starch at early stages of maturation and progressively degrade starch to increase sweetness etc. during ripening [28–30]. Starch granules are formed in amyloplasts within cells and have a size from 1 to 100 ␮m. Laser light of 632.8 nm is scattered on starch granules. Starch does not move around together with organelles, however the cyclosis presumably causes some vibrations of the granules. Thus, apart from other moving organelles, starch granules would give many additional non-stationary scattering centers. In result, more starch particles means higher apparent biospeckle activity. Next it has been studied that less chlorophyll content causes higher apparent biospeckle activity [14]. As the fruit ripens, the incoming light penetrates more deeply into it due to the degradation of chlorophyll (which has an absorption peak close to the wavelength of the He–Ne laser used). Next the respiration plays a major role in the postharvest life of fresh commodities because it reflects the metabolic activity of the tissue that includes the loss of substrate, the synthesis of new compounds, and the release of heat energy. It is the oxidative breakdown of complex substrate molecules normally present in plant cells, such as starch, sugars, and organic acids, to simpler molecules such as CO2 and H2 O. Generally there is an inverse relationship between respiration rates and postharvest-life of fresh commodities. The higher the respiration rate, shorter postharvest-life, the commodity usually is [31]. Process of respiration within the fresh commodities results in loss of substrate molecules (e.g. sugar, starch content), loss of chlorophyll, etc. During the development of the flesh of a fruit, nutrients are deposited as starch which during the ripening process is transformed into sugars. The progression of the ripening process leads to decreasing starch levels. Thus the biospeckle activities of the fresh commodities should decrease with the decrease in their rates of respiration. 5. Conclusion The formalism of IM algorithm represents the biospeckle activity by a non-dimensional number, where the high activity is presented by higher values and lower activity by lower values. The study using two methods cross correlation coefficients and inertia moment, of the different fruits leads to the following conclusion: 1. The cross-correlation coefficient of biospeckle patterns decreases faster when the fruits are fresher. The cross correlation coefficient continues to decrease faster for first few days and then the decrease is slowest in tomato and apple than in pear. 2. The biospeckle activity decreases with aging of the fruits. The decrease is more in pear but relatively lower in the case of apple and tomato as predicted by cross correlation technique.

2185

3. The decrease in biospeckle activities can be explained by their respiration rates. The respiration rate is high in case of pear and low for apple and tomato at temperatures (20–25 ◦ C) [31]. So as can be seen from Table 2, the biospeckle activity for apple, pear and tomato decreases according to their individual rates of decrease of respiration. 4. By the comparative study between the two methods it is found that IM is more reliable to predict bioactivity levels in fruits in comparison to that of cross-correlation technique as IM measures bioactivity levels directly. 5. The bioactivity is higher for pear in comparison to the apple and tomato as predicted by IM methods. Acknowledgment We are sincerely thankful to DST, New Delhi for giving financial assistance of Rs. 39.902 lakhs as sponsored project no. “SR/S2/LOP07/2005 dated 24/10/2007” without which it would not have been possible to perform this works at all. References [1] J.A. Abbott, Quality measurement of fruits and vegetables, Postharvest Biol. Technol. 15 (1999) 207–225. [2] Z. Zude-Sasse, I. Truppel, B. Herold, An approach to non-destructive apple fruit chlorophyll determination, Postharvest Biol. Technol. 25 (2002) 123–133. [3] P.E. Zerbini, M. Grassi, R. Cubeddu, A. Pifferi, A. Torricelli, Time-resolved reflectance spectroscopy can detect internal defects in pears, Acta Hortic. 599 (2003) 359–365. [4] Y. Peng, R. Lu, Analysis of spatially resolved hyperspectral scattering images for assessing apple fruit firmness and soluble solids content, Postharvest Biol. Technol. 48 (2008) 52–62. [5] F. Firtha, A. Fekete, T. Kaszab, B. Gillay, M. Nogula-Nagy, Z. Kovács, D.B. Kantor, Methods for improving image quality and reducing data load of NIR hyperspectral images, Sensors 8 (2008) 3287–3298. [6] Z. Qing, B. Ji, M. Zude, Predicting soluble solid content and firmness in apple fruit by means of laser light backscattering image analysis, J. Food Eng. 82 (2007) 58–67. [7] L. Baranyai, M. Zude, Analysis of laser light propagation in kiwifruit using backscattering imaging and Monte Carlo simulation, Comput. Electron. Agri. 69 (2009) 33–39. [8] W.B. Herppich, Application potential of chlorophyll fluorescence imaging analysis in horticultural research, in: Proceedings of the 6th International Symposium: Fruit, Nut and Vegetable Production Engineering, 11–19 September, Potsdam, Germany, 2001, pp. 609–614. [9] W.B. Herppich, M. Linke, S. Landahl, A. Gzik, Preharvest and postharvest responses of radish to reduced water supply during growth, Acta Hortic. 553 (2001) 89–90. [10] E. Bauriegel, A. Giebel, W.B. Herppich, Hyperspectral and chlorophyll fluorescence imaging to analyse the impact of Fusarium culmorum on the photosynthetic integrity of infected wheat ears, Sensors 11 (2011) 3765–3779. [11] B.M. Nicolai, K. Beullens, E. Bobelyn, A. Peirs, W. Saeys, K.I. Theron, J. Lammertyna, Nondestructive measurement of fruit and vegetable quality by means of NIR spectroscopy: a review, Postharvest Biol. Technol. 46 (2007) 99–118. [12] R.A. Braga, L. Dupuy, M. Pasqual, R.R. Cardosos, Live biospeckle laser imaging of root tissues, Eur. Biophys. J. 38 (2009) 679–686. [13] R.A. Braga JR, W. Silva, T. Safadi, C. Nobre, Time history speckle pattern under statistical view, Opt. Commun. 281 (9) (2008) 2443–2448. [14] G.G. Romero, C.C. Martinez, E.E. Alanis, G.A. Salazar, V.G. Broglia, L. Alvarez, Biospeckle activity applied to the assessment of tomato fruit ripening, Biosyst. Eng. 103 (2009) 116–119. [15] A. Zdunek, J. Cybulska, Relation of biospeckle activity with quality attributes of apples, Sensors 11 (2011) 6317–6327. [16] J.D. Briers, A.F. Fercher, Retinal blood-flow visualization by means of laser speckle photography, Invest. Ophthalmol. Vis. Sci. 22 (1982) 255–259. [17] R.A. Braga, I.M. DalFabbro, F.M. Borem, G. Rabelo, R. Arizaga, H.J. Rabal, M. Trivi, Assessment of seed viability by laser speckle techniques, Biosyst. Eng. 86 (2003) 287–294. [18] G.H. Sendra, R. Arizaga, H.J. Rabal, M. Trivi, Decomposition of biospeckle images in temporary spectral bands, Opt. Lett. 30 (2005) 1641–1643. [19] J.A. Pomarico, H.O. DiRocco, L. Alvarez, C. Lanusse, L. Mottier, C. Saumell, R. Arizaga, H.J. Rabal, M. Trivi, Speckle interferometry applied to pharmacodynamic studies: evaluation of parasite motility, Eur. Biophys. J. 33 (2004) 694–699. [20] R.A. Braga, G.F. Rabelo, L.R. Granato, E.F. Santos, J.C. Machado, R. Arizaga, H.J. Rabal, M. Trivi, Detection of fungi in beans by the laser biospeckle technique, Biosyst. Eng. 91 (2005) 465–469. [21] M. Pajuelo, G. Baldwin, H.J. Rabal, N. Cap, R. Arizaga, M. Trivi, Bio-speckle assessment of bruising in fruits, Opt. Lasers Eng. 40 (2003) 13–24.

2186

M.D.Z. Ansari, A.K. Nirala / Optik 124 (2013) 2180–2186

[22] Y. Tao, Z. Wen, Machine Vision Techniques for Defect Inspection on Fruit Packing Lines. Advances in Bioprocessing Engineering, World Sci. Publ. Co. Ltd., 2002. [23] M. Sjödahl, Some recent advances in electronic speckle photography, Opt. Laser Eng. 29 (1998) 125–144. [24] A. Zdunek, L.I. Muravsky, L. Frankevych, K. Konstankiewicz, New nondestructive method based on spatial–temporal speckle correlation technique for evaluation of apples quality during shelf-life, Int. Agrophys. 21 (2007) 305–310. [25] Md. Zaheer Ansari, A.K. Nirala, Assessment of bio-activity using the methods of inertia moment and absolute value of the differences, Optik, http://dx.doi.org/10.1016/j.ijleo.2011.12.013, in press. [26] A. Oulamara, G. Tribillon, J. Duvernoy, Biological activity measurement on botanical specimen surfaces using a temporal decorrelation effect of laser speckle, J. Mod. Opt. 36 (1989) 165–179.

[27] R. Arizaga, M. Trivi, H.J. Rabal, Speckle time evolution characterization by the co-occurrence matrix analysis, Opt. Lasers Technol. 31 (1999) 163–169. [28] I.J. Warrington, T.A. Fulton, E.A. Halligan, H.N. de Silva, Apple fruit growth and maturity affected by early season temperatures, J. Am. Soc. Hort. Sci. 124 (1999) 468–477. [29] M.K. Ernst, G. Matitschka, N.J. Chatterton, P.A. Harrison, A quantitative histochemical procedure for measurement of starch in apple fruit, Histochem. J. 31 (1999) 705–710. [30] D.G. Stevenson, P.A. Domoto, J. Jane, Structures and functional properties of apple (Malus domestica Borkh) fruit starch, Carbohydr. Polym. 63 (2006) 432–441. [31] J.B. Biale, R.E. Young, Respiration and ripening in fruits—retrospect and prospect, in: J. Friend, M.J.C. Rhodes (Eds.), Recent Advances in the Biochemistry of Fruits and Vegetables, Academic Press, NY, 1981, pp. 1–39.